Most operations end up running on a patchwork: a CRM used as a database, a spreadsheet nobody dares change, three systems that don't speak, and a person whose actual job is retyping between them. It works — until volume arrives. Then the patchwork is the ceiling.
Data re-keyed between systems, approvals chased over email, reports rebuilt from scratch every month. Effort scales with volume because nothing runs end to end.
Finance, operations and the customer-facing team each hold a different version of the same number. Decisions get made on whichever sheet is open.
An inherited system with no documentation, no tests, and nobody left who wrote it. Every change carries risk, so changes stop being made.
Six delivery lines kept in a single team, because most of what we are asked to build sits across several of them at once.
Marketing sites, full-stack web applications and mobile builds — fast, accessible and cheap to change afterwards.
CRM, ERP, billing and SaaS platforms built around your actual workflow instead of configured around someone else's product.
AI/ML solutions, workflow automation and chat agents that take repetitive work off the day — then get measured in hours returned.
Storefronts and the operational systems behind them — payments, inventory, orders and CRM synchronised rather than reconciled by hand.
Admin panels and reporting layers for the point where a spreadsheet stops being able to carry the operation.
UI/UX and brand work, plus the search, paid and content system that puts a finished product in front of people.
Chosen because we maintain them in production ourselves, not because they trended this quarter.
Client case studies are still being cleared for publication. In the meantime the honest evidence is the software we build, operate and carry the cost of ourselves.
A modular ERP for institutions and mid-sized operators covering people, scheduling, finance and reporting as separate modules rather than one monolith — so a team can adopt the part that hurts most first.
Our education and training vertical: internships, technical courses and structured training run against real engineering work. Over 5,000 people have been trained through SkillTank internships and courses.
Why this matters to you: running our own products means we live with our own architectural decisions. When we recommend a stack, a schema or a deployment model, it is one we already maintain in production.
Payment rails, tax-aware reporting and the messaging channels your customers actually answer are configured for your market, not retrofitted from a template. We run the same stack in production on our own products, so the tools we recommend are ones we already maintain.
Payment rails, tax-aware reporting and the messaging channel your customers actually answer, configured for the market you sell into rather than retrofitted from a template.
Deep, repeated work in Google Workspace and Apps Script, Razorpay, and the WhatsApp Business API — the integrations operations actually run on.
SkillTank has put over 5,000 people through structured internships and courses. Our engineering standards stay written down because they have to be teachable.
The engineer who scoped your project writes it, and the person answering your question has read the code — a discipline we intend to keep as we scale.
The same five stages on every engagement, whether the brief is a platform, an integration or an automation. Nothing here is invented for a sales conversation.
We follow the work as it runs today and write it down — where the hours go, where data is re-keyed, where errors enter.
Deliverables, integration points, phases and an explicit out-of-scope list, agreed in writing before any code exists.
Working software in reviewable increments, with a running environment you can open at any point during the build.
Type checks, automated tests and a manual pass on the paths that matter. Releases are instrumented and reversible by design.
Architecture notes and runbooks ship with the build, then we keep reading the numbers with you and tightening the system.
We build from Chennai and Chidambaram and work across time zones. Remote-first and asynchronous by default, with the working-hour overlap agreed per engagement rather than assumed.
Markets we serve
Rotating view — markers show the markets listed alongside.
Engineers, designers and SkillTank trainees working out of Chennai and Chidambaram, Tamil Nadu.
Status, milestones, invoices and documents for every system we run for you.
Describe the process that costs you most and we will tell you what it would take to fix — including when the answer is that software is not the right lever.
Most engagements begin on one line and pick up a second as the system grows. Nothing here is a separate department with its own handoff.
Enterprise applications, SaaS platforms and internal systems, built around how a team actually works rather than around a product category.
Software built to one operation: its workflow, its permissions model, its reporting — instead of a generic feature set you configure around.
Multi-user platforms and portals for customers, staff or partners, with the authentication, roles and audit trail real usage demands.
Multi-tenant architecture, billing, provisioning and the operational tooling a product needs before it can carry paying customers.
Systems connected so information is entered once. API design, third-party integration, and reconciliation logic that keeps two sources agreeing.
Applications that reach staff and customers on the device they actually use, sharing a codebase with the web product where that makes sense.
Inherited systems documented, tested and made changeable again — usually incrementally, because a full rewrite is rarely the cheapest route.
We connect the tools you already pay for and take the repetitive steps out of the working day, then measure the hours it gives back. This is the line we are asked for most.
Classification, extraction, matching and scoring applied to a specific operational problem — resume parsing for recruiters being a representative example.
Notifications, status updates, reminders, qualification and support handled on the channel your customers already answer.
Apps Script, Sheets and Drive turned from a manual filing system into a running process — a stack a great many operations sit on.
Collection, reconciliation, subscription handling and payout workflows connected to your books rather than checked by hand.
Multi-step processes that currently move by hand — intake, routing, approval, dispatch — rebuilt to run without someone shepherding each stage.
Invoices, forms, resumes and contracts read, extracted and validated, then written into the system that needs the data.
Ticket triage, drafted responses, lead assignment and pipeline hygiene — with escalation rules so uncertain cases reach a human.
Recurring reports and cross-system matching produced on a schedule, with exceptions surfaced rather than buried.
Assistants grounded in your own documentation and records, scoped by role and logged so answers are traceable.
Where we say no: automation multiplies whatever it is pointed at. If the underlying process is broken, automating it produces bad outcomes faster. In that case we fix the process first and quote for less work.
The admin surfaces and reporting layers a growing operation stops being able to run without — usually the first thing a spreadsheet stops being able to carry.
Operations consoles for the team running the business day to day, with the permissions and audit trail that come with real responsibility.
Live operational reporting built on the records you already keep, so the number on screen is the number in the system.
Configuration, extension or replacement of the systems of record — including custom modules where an off-the-shelf product stops fitting.
The plumbing underneath the dashboard: collection, cleaning and scheduled movement between systems.
Customer, partner and staff portals that remove a category of email from the working week.
Shared services, authentication and the internal tooling that stops every new build starting from zero.
Design and engineering sit in the same team. Layouts are drawn against real content and real data, then built by the people who drew them, so what ships is what was agreed.
Production builds with readable code, sensible performance budgets and a content model your team can actually operate.
Managed hosting sized to real traffic, with deployment pipelines, backups and monitoring configured rather than assumed.
Ongoing updates, dependency hygiene and incident response. Measured on uptime, load time, and how cheap the next change is to make.
Interface design and prototypes tested before a build starts, because the cheapest time to find a usability problem is before it is code.
Contrast, keyboard operation and semantics treated as build requirements; identity work specified so engineering can implement it.
Storefront and checkout builds wired to Razorpay and UPI, with order, inventory and CRM synchronisation behind them.
For teams deciding what to build, what to buy, and what to stop doing — plus the marketing work that puts a finished system in front of people.
A written map of how work runs today, with the cost of each bottleneck quantified before any technology is chosen.
Independent advice on stack, platform and build-versus-buy, with the operating costs written down alongside the licence fees.
Sequencing a multi-year change so each phase pays for itself rather than requiring faith until the end.
SEO, paid acquisition and landing pages run as one measurable system with spend tied back to pipeline.
Editorial and lifecycle messaging maintained on a cadence, aimed at the questions buyers actually ask.
AI-assisted launch films, social cuts and always-on content, produced fast enough to keep a channel fed.
We work as the delivery team behind other agencies, studios and consultancies — white-labelled build, automation and maintenance under your brand and on your schedule.
Same scoping discipline, same review practice, same handover notes — delivered to your client on your timeline and presented as your team.
Extra capacity when a quarter is oversold, or specific depth — backend, cloud, data, AI — your team does not carry in-house.
We can take an isolated component with a clear interface while your team keeps the rest, rather than requiring the whole project.
A standing arrangement with agreed rates, a named engineering contact and priority on capacity when your quarter gets busy.
Send work our way that does not fit your studio. We will tell you honestly whether we can deliver it, and referral terms are agreed in writing beforehand.
We work under your NDA rather than insisting on ours, and we do not publish anything from a partner engagement.
| Shape | Use it when | What you receive |
|---|---|---|
| Discovery | The problem is clear, the solution is not | Written process map, technical options with trade-offs, phased estimate |
| Build | Scope is agreed and ready to ship | Working software in increments, tests, documentation, handover |
| Programme | Several linked systems across service lines | Sequenced roadmap, staged delivery, measurement against agreed baselines |
| Retained | Live systems needing support and iteration | Maintenance, monitoring, an agreed change allowance and priority response |
| Partnership | You are an agency needing a build team | White-labelled delivery under your brand |
On pricing: we quote after discovery rather than from a rate card, because an estimate given before the process is mapped is wrong in both directions. Ask on the first call and we will give you an honest range for work of a similar shape.
Not sure a build is justified yet? The payback calculator gives you the annual cost of the manual version and says plainly when automating it is not worth the money.
Describe the symptom. Working out which discipline it belongs to is our job, not yours.
Pick whichever is closest to what is actually annoying you. You do not need to know what the solution is called — working that out is our job, not yours.
Worth establishing before either of us spends a call on it.
Two systems developed and operated in-house. They are where architecture and delivery ideas get tested at our own expense before they appear in a client project — and they are the clearest evidence of how we build.
Kaizen One is not a product off the shelf. It is the whole Kaizen module library — ERP, CRM, HRMS, finance, inventory, analytics and the rest — assembled and rewritten to fit how one particular business actually runs. Where a process has no equivalent in any existing module, we build that module. The result is a single system carrying your workflow, your terminology and your reports, rather than a standard product you have to bend your operation around.
Start from the twenty-six, take the modules the operation actually uses and leave the rest. There is no bundle you have to buy around a feature you wanted.
Screens, fields, approvals, documents, workflows and reports are rewritten to your process. Configuration flags only reach so far; past that point we change the code.
Where an operation does something none of the twenty-six covers, that becomes a new module inside the same system, on the same permissions and reporting layer as the rest.
However many modules run, there is one login, one audit trail and one set of numbers. Custom does not mean a set of separate tools bolted together.
Source, schema and deployment documentation transfer with the build. You are not renting access to something only we can change.
The point of a custom system is that it keeps fitting. Changes are scoped and shipped in cycles rather than saved up for a rewrite in year three.
Each is a finished system for a specific operation, not a template configured after the sale. Run one on its own and nothing else is required. Run several and they share one login, one permissions model and one reporting layer, so the numbers reconcile without anybody exporting a spreadsheet.
Enterprise & business management
View productCustomer & sales management
View productHuman resources & payroll
View productAccounting, billing & expenses
View productProject & task management
View productInventory & warehouse management
View productBI, dashboards & reporting
View productAI agents & business automation
View productSchool education management
View productCollege & university ERP
View productHospital management system
View productClinic management
View productPatient-facing healthcare platform
View productManufacturing ERP
View productProduction & shop-floor management
View productReal estate CRM & management
View productConstruction project management
View productRetail ERP
View productPoint of sale
View productHotel management
View productRestaurant management
View productLogistics management
View productFleet & transportation management
View productInstitutional management
View productAshram & spiritual institution management
View productNGO & nonprofit management
View productNo products in that group.
All twenty-six, in full. Use the filter above to narrow this list to a single sector.
One ledger behind purchase, stock, sales and finance, so a transaction is entered once and every report reads the same record. Modules switch on independently, which means a team can start with the function that hurts most rather than migrating everything over one weekend.
No single cutover. One department moves one function across, proves it, and the rest follows from there.
Role-based access with an audit trail on record changes, because in an ERP the question is rarely what the value is and usually who changed it and when.
Reports read the operational records directly rather than a nightly export, so the number on the dashboard is the number in the system.
Leads, quotations, follow-ups and orders held against one customer record, so nobody has to ask which spreadsheet is current. The pipeline is tracked by stage, and the history of every conversation stays attached to the account.
Enquiries, quotes, orders and complaints on the same account, visible to whoever picks up the call.
Deals move through stages you define, with ageing, so a stalled deal surfaces instead of quietly sitting.
Reminders, call logs and WhatsApp or email touchpoints scheduled against the record rather than somebody's memory.
Attendance through to payslip in one system, with the statutory registers generated from the same data instead of rebuilt in a spreadsheet every month.
Biometric, mobile or manual capture, leave balances that accrue by policy, and approvals with a trail.
Earnings, deductions, arrears and reimbursements calculated to your structure, with payslips issued from the run itself.
PF, ESI, professional tax and TDS registers produced from the payroll that was actually run, not re-entered.
Invoicing, ledgers, receivables and expense claims in one place, with tax-ready data falling out of the transactions rather than being reconstructed at quarter end.
Raise, send, track and reconcile, with ageing on what is outstanding and a record of who has been chased.
Double-entry underneath, so trial balance, profit and loss and balance sheet all come out of the same postings.
Claims submitted with receipts, approved against policy, and posted to the right head without re-keying.
Tasks, milestones and time booked against a plan, so effort spent can be compared with effort quoted while the project is still running and something can still be done about it.
Milestones, task owners and dependencies, with slippage visible before the deadline rather than after it.
Hours booked to task, which gives you both an invoice line and an honest number for the next estimate.
Cost and billing tracked per project, so margin is a live figure instead of a post-mortem.
Stock by location, bin and batch with every movement recorded, so the physical count and the system agree far more often than they otherwise would.
Bin-level placement with put-away and picking, so stock gets found rather than searched for.
Traceability by batch, serial and expiry, which is what matters on the day something has to be recalled.
Inward, transfer, issue and dispatch all documented, with variance flagged at the point it happens.
Dashboards built on the operational tables themselves, refreshed live rather than assembled from exports, plus scheduled reports that arrive without anyone remembering to run them.
Charts read the same records the system writes, which removes the argument about whose number is correct.
Reports go out on a timetable to the people who need them, in the format they already work in.
Every figure opens to the transactions behind it, so a surprising number can be checked in seconds.
Agents that take the repetitive edge of the business - capturing leads, answering routine WhatsApp queries, reading documents - and hand anything unusual to a person with the context attached.
Enquiries from web, WhatsApp and calls captured, qualified and written into the CRM without manual entry.
Invoices, purchase orders and forms read and posted, with low-confidence cases queued for a human to check.
The agent knows what it does not know. Anything outside its brief goes to a named person, not into a void.
Admissions through to results for a school, with attendance, timetable and fee collection all running off one student record.
One record per student across every year, so history is not rebuilt at each promotion.
Period-wise attendance against the published timetable, with absence visible to the class teacher the same day.
Structures, concessions, instalments and receipts, with an ageing list of exactly what is outstanding.
Departments, semesters, electives, internal marks and examinations for higher education, with results processed from the same records that hold attendance.
Course registration, credits and electives per programme, with eligibility checked at the point of registration.
Internal and external marks, moderation, grade points and transcripts through one pipeline.
Each department sees its own load, attendance and results without raising a request with the office.
Outpatient and inpatient flows in one system - registration, consultation, wards, theatre, pharmacy, diagnostics and billing - all posting against the same patient record.
One registration serves both, so a patient admitted from outpatient does not get entered a second time.
Bed allocation, procedure scheduling and drug issue recorded against the case, and billed from it.
Charges accrue as care is given, so the discharge bill is assembled rather than reconstructed at the counter.
Appointments, patient history, prescriptions and counter billing for a single clinic or a small chain, without the weight of a hospital system behind it.
Slots by doctor, with reminders to the patient and a live queue view for the front desk.
Past visits, notes and prescriptions on one screen, printable or sent straight to the patient.
Consultation, procedure and pharmacy on one bill, with the day's collection closed off at the counter.
The side the patient sees: booking, records access and teleconsultation, connected to whichever clinical system holds the data underneath.
Self-service appointments across locations and practitioners, with confirmation and reminders.
Reports, prescriptions and visit history available to the patient without a phone call to reception.
Video consultation with the note and prescription written into the same record as an in-person visit.
Bill of materials, work orders, job work and costing, so the cost of a finished item is built from what was actually consumed rather than a standard rate applied afterwards.
Multi-level bills, work orders raised against demand, and material issued specifically to the order.
Material sent out, processed and received back, tracked with the challan and the loss recorded at each stage.
Material, labour and overhead accumulated per process, giving a cost per unit you can defend in a meeting.
What actually happens on the floor: routing, machine loading, operator output and wastage, captured at the station rather than written up at the end of the shift.
Operations sequenced across machines with capacity visible, so the plan resembles the floor.
Production and rejection entered where the work happens, which is what keeps the number current.
Stoppage reasons and scrap recorded, so recurring causes become countable instead of anecdotal.
Unit inventory, enquiries, site visits and bookings on one pipeline, from the first call through to agreement and payment schedule.
Availability, hold and booking status per unit, so two sales people cannot sell the same flat.
Enquiry source, follow-up and visit scheduling, with conversion visible per channel rather than in aggregate.
Agreement value, milestone-linked demand and receipts all tracked against the unit.
Site-level costing and progress billing, with labour, material and subcontractor cost reconciled against the estimate while the work is still going on.
Budgeted quantity and rate compared with what has been consumed, per activity and per site.
Measurement books, certified quantities and running account bills raised from the same measurements.
Indent, issue and return at site, and labour deployment by contractor with attendance.
Purchase, pricing, stock and settlement across stores, so head office is looking at the same numbers each counter is transacting on.
Stock, transfer and reorder per store, with a consolidated view for the buying team.
Price lists, discounts and offers applied at the counter and reported back at the centre.
Cash, card and digital collections tallied per till and per day, with variance flagged the same evening.
Counter billing that keeps working when the connection does not, syncing back when it returns, with shift-wise cash reconciliation at the end of it.
Billing continues without the network. Transactions queue locally and sync the moment it is back.
Barcode, keyboard shortcut and touch input, built around queue speed rather than around a form.
Opening float, collections and closing cash per shift, with the difference recorded and owned by someone.
Reservations, front desk, housekeeping and folios, with rates and availability pushed out to the booking channels from one place.
Booking, check-in, charges and settlement on one guest folio, including extras posted from the outlets.
Room status live between the desk and the floor, so a cleaned room becomes sellable immediately.
Rate and inventory pushed to the OTAs and pulled back, which removes most of the overbooking manual updates cause.
Table and delivery orders, kitchen routing and recipe-level costing, so the margin on a dish is a known figure rather than an assumed one.
Dine-in, takeaway and aggregator orders on one screen, with table state visible to the floor.
Order tickets sent to the right section, with preparation time tracked against what was promised.
Consumption deducted by recipe, giving food cost per dish and variance against theoretical usage.
Consignment booking through to delivery: hubs, manifests, freight billing and proof of delivery, with the status the customer sees taken from the same record.
Consignment note, hub scan, delivery run and signed proof, all against one docket.
Rate contracts by lane, weight and customer, billed from the consignments that actually moved.
Tracking the consignee can check themselves, which takes most of the phone traffic off the branch.
Vehicles, drivers, trips and running cost, so cost per kilometre is a figure that comes out of records rather than off the top of someone's head.
Trip, distance, halt and expense captured per vehicle, with driver allowance calculated from it.
Fuel entries against odometer, service schedules by date or distance, and breakdown history per vehicle.
Insurance, fitness, permit and licence expiries tracked, with warning well before they lapse.
Members, programmes, facilities and receipts for a body that does not fit the shape of a school, a hospital or a company - associations, trusts, societies and clubs.
Membership categories, renewals and dues, with receipts and reminders handled from the record.
Event scheduling, room and facility booking, and attendance wherever it needs recording.
Collections posted to heads that reconcile with the accounts, which is what an audit will ask for.
The particular record-keeping of an ashram or temple trust: devotees, donations, accommodation, annadhanam, seva bookings and event rosters.
Devotee records with seva and archana bookings, receipts issued at the counter or online.
Donation heads, in-kind entries and receipts carrying the certification a donor needs at year end.
Room allotment for visitors and daily annadhanam counts, with the provisioning that follows from them.
Donors, grants, beneficiaries and field activity in one place, so a utilisation report is produced from records rather than assembled from memory the week before a deadline.
Commitments, receipts and restrictions per donor, with what each grant may be spent on held against it.
Beneficiary records and field activity captured on the ground, including where there is no signal.
Spend mapped to grant and to outcome, in the shape the funder and statutory reporting expect.
No products in that group.
All twenty-six are available white-label. Partners sell them under their own brand while we build, host and support the software behind it — see the SaaS Partner Program. Where an operation does not fit a shelf product, we build to the process instead.
SkillTank is the education and training vertical of KaizenSpark Tech — internships, technical courses and structured training, all run against real engineering work rather than set exercises. Trainees work inside live codebases, with code review and pairing as the teaching mechanism. Over 5,000 people have been trained through SkillTank internships and courses.
Work that gets reviewed and merged, not a shadowing exercise or a sandbox task set aside from the real thing.
Teaching happens in code review and at the keyboard rather than in lectures, which is how standards actually transfer.
The course material is our own engineering documentation. Keeping it teachable is what keeps it current.
Kaizen One and the twenty-six are ours, and SkillTank is our education vertical. The majority of our engineering work is built to a client brief — internal platforms, integrations and automation shaped around an operation that already exists rather than sold off a shelf. If neither product above matches what you need, that is the usual case.
We will walk you through the product, the roadmap, and what it would take to fit it to your operation.
Kaizen means continuous improvement. The engineering translation is unglamorous: ship something narrow, instrument it, read the numbers, and keep tightening. Most builds skip the last part, which is why most builds decay.
The same order whether the brief is a platform, an integration or an automation. Nothing in it is invented for a sales conversation.
You can run part of stage one yourself. The readiness audit asks the ten questions we would ask on a first call, and tells you where a process is not yet ready to automate.
We sit with the people doing the job and follow the work step by step: where the hours go, where data gets re-keyed, where errors enter, and which spreadsheet everyone actually trusts. The deliverable is a written map of the current state. It is deliberately produced before any technology is discussed, because a solution chosen before the process is understood is a guess with a budget attached.
Deliverables, the systems it has to talk to, the phases, and what is explicitly out of scope — agreed in writing before code exists. Scope changes are normal and welcome; silent scope changes are not. This is also where we tell you if a smaller piece of work solves the problem, which happens more often than it is convenient to admit.
Working software in short cycles with a running environment you can open at any point, rather than a long build behind a status report. Each cycle ends in a demo. Nothing is called done because it was written — it is done when someone outside the pair that built it has used it and the tests pass.
Type checks, automated tests and a manual pass on the paths that matter run before anything reaches production. Releases are reversible. We agree the measurements up front and instrument what we ship — time per task, error rate, hours returned, queue length — because without a baseline captured before launch there is nothing honest to compare against afterwards.
Architecture notes, environment setup and runbooks are written as part of delivery, not sold afterwards. The test is whether an engineer who was not on the project can operate it without a meeting. Then real usage data starts arriving, and the next increment gets decided by evidence rather than by the original plan.
Six technical disciplines, kept in one team because most of the problems we are asked to solve sit across several of them at once.
Enterprise applications, SaaS platforms and internal tools built around how a team works rather than a generic product category.
Production builds, hosting and long-term maintenance, measured on load time, uptime and the cost of the next change.
Pipelines, reporting and dashboards that turn records a business already keeps into numbers someone can act on this week.
Infrastructure, deployment pipelines and environments sized to real load, with operating costs written down before we build.
Access control, dependency hygiene, secrets handling and review practice built into delivery instead of bolted on at the end.
Connected devices, telemetry, and the software that collects, stores and makes sense of what those devices report.
Operational rather than aspirational. Each one has a cost we actually pay.
A narrow system this quarter, improved every month after, beats one large release nobody has used. Shorter first release, longer useful life.
Claims come with a number and a method or they do not go in the report. Where something cannot be measured, we say that rather than round it up into a result.
Architecture decisions, handover notes and runbooks are part of delivery. Decisions live in documents rather than in someone's memory.
All code goes through review regardless of who wrote it. It is how standards stay consistent as the team grows, and how people here learn fastest.
We will tell you when software is not the answer or when a cheaper process change gets the same outcome. Losing a project in scoping costs less than delivering the wrong one.
The months after release decide whether a system compounds or decays. We plan for that window instead of treating handover as the end.
Usually not. A large share of the value is in connecting systems you already pay for — Google Workspace, your CRM, Razorpay, WhatsApp. We recommend replacing a tool only when keeping it costs more than the migration would.
You do. Data stays in your accounts wherever the architecture allows, access is scoped to what each system needs, and we document data handling in writing before a build starts.
Yes, and it is often the better arrangement. We can specify and hand off, pair on delivery, or take a defined slice with a clear interface while your team keeps the rest.
Then the design was wrong. Anything customer-facing ships with confidence thresholds, human review on uncertain cases, and a logged trail so a failure is diagnosable rather than mysterious.
Yes. Our defaults — UPI and Razorpay, GST-aware reporting, WhatsApp as a primary channel, IST hours — run deep because we use them daily, and none of them prevent us building on whatever stack your market runs on.
The fair answer is to test the method rather than take the history on faith. Ask how we scope, how we estimate, how we test and what handover contains — and ask us to walk you through the architecture of the Kaizen products, which we build and operate ourselves.
It is the cheapest stage of the engagement and the one that determines whether the rest is worth doing.
KaizenSpark Tech Private Limited was incorporated in November 2024 and is registered in Tamil Nadu. We are an engineering company working across custom software, AI and IT services — and we are growing deliberately. In under two years we have delivered for more than fifty companies, trained over five thousand people, and taken on work across eight markets — and we are putting the engineering discipline in place now that a far larger company will need later.
Every engagement runs through the same review practice and the same written standards. That is what lets us take on enterprise work today and add people without the quality falling away tomorrow.
KaizenSpark Tech Private Limited, registered in Tamil Nadu, India. Operating from Chennai and Chidambaram, Tamil Nadu, and serving eight markets.
A young company is a fair thing to ask about. The answer is that documentation, runbooks and environment setup are delivered as part of every build specifically so you are not dependent on our availability — an engineer who never worked on the project should be able to operate it from the documents. How that is handled.
Client work across software builds, automation, dashboards and web — most of it custom rather than productised.
Through SkillTank internships and courses. It is how we grow the bench we hire from and why our standards stay written down.
Small enough that you will know who is building your system, and who to call when you need them.
Company direction, client scoping, and the final call on what we take on and what we turn down. B.Tech in Artificial Intelligence & Data Science. Writes on automation economics and where AI spend actually returns.
Architecture, delivery standards and the technical review every build passes before it ships. Owns the handover standard — runbooks, documentation and the credentials that transfer with the code.
Application and integration work — payment rails, reconciliation, Apps Script and the connective code between systems that were never designed to talk to each other.
Hiring, onboarding and SkillTank by KaizenSpark Tech — the internship and training programme the engineering bench is grown from.
The people are named above. This is the structural version: what each seat covers, and which of them one person is currently carrying. Today people often cover more than one of these. Each seat is being built out into a team as the company grows.
Company direction, client scoping, and the final call on what we take on and what we turn down.
Platform architecture and delivery. Owns technical review, estimates, and how a build is sequenced.
Application and API development across the custom software and SaaS work.
Deployment pipelines, environments, cost sizing, and the operational side of what we ship.
Pipelines, reporting, and the models behind our AI and automation work.
Interface design, prototypes, and the usability testing that happens before a build starts.
Specialist work outside these seats — security review, IoT firmware, video production — is brought in per project and named in the statement of work before it starts.
Work is organised in short cycles with a demonstration at the end of each. That rhythm is the reason estimates get more accurate the longer an engagement runs.
Remote-friendly, and decisions get written down rather than living in someone's memory. If it was not recorded, it was not decided.
A prototype or a measurement wins the argument, not seniority. Slower in the moment and considerably faster over a project.
New tools, model providers and architectural patterns go into our own Kaizen products and SkillTank before they go near a client project. When an approach fails there it costs us a sprint rather than costing you a release.
We hire deliberately, and mostly from people we have already worked with. The main route in is SkillTank, our internship and course programme, which has trained over 5,000 people: trainees join a supervised track inside our own codebases, with review and pairing as the teaching method. It is deliberately not a shadowing exercise — you write code that gets reviewed and merged.
We publish roles as they open rather than keeping a standing list. If one of the seats above describes what you do, or if you want a place on the next SkillTank intake, write to us with something you have built and what you would change about it now. That last part is the bit we read closely.
We would rather lose a project during scoping than deliver the wrong one — and we will tell you when what you are asking for is not worth building.
Tell us where the hours go, or what keeps breaking. We will tell you whether software is the right answer — and if it is, what it would take.
Most clients start with discovery and move into a build. Programme and retained arrangements follow from there. You do not have to pick correctly — that is what the first call is for.
Interviews with the people doing the work, a written map of the current process, bottlenecks identified and costed, technical options with trade-offs, and a phased estimate for the build.
Pick this if the problem is clear but the solution is not.
Priced per phase from the discovery estimate. Working software in reviewable increments, a running environment you can open at any point, tests and pre-release checks, architecture notes and runbooks, and team training at handover.
Pick this if scope is agreed and ready to ship.
Rolling, with an agreed notice period. Maintenance, iteration and support on systems that are already live and earning.
Pick this if something is running and needs to keep improving.
Several builds sequenced over a longer horizon, where the pieces depend on each other and the order matters as much as the work.
Pick this if this is a roadmap rather than a project.
Every one of these includes the written scope, the out-of-scope list, code review, runbooks and IP transfer. Those are not line items.
These are the three options actually on your desk. Here is the honest shape of each, including where we are the wrong answer.
| Hiring in-house | Us | A larger agency | |
|---|---|---|---|
| Time to start | Months. Sourcing, notice periods, then ramp-up. | Weeks. Discovery can usually start inside one. | Weeks, but often behind their existing queue. |
| What you carry | Salary, equipment, benefits and the management of one more person — whether or not there is work that week. | A fixed scope at a fixed price. When the build ends, the cost ends. | A retainer, usually with a minimum term. |
| Depth of experience | One person's, at whatever level you can afford. | Patterns from many builds, but a small team — we are not everywhere at once. | Broad, though the people who pitched may not be the people who build. |
| Who writes the code | The person you hired. | The engineer who scoped it. The person answering your question has read it. | Frequently a different team from the one you met. |
| If they leave | The knowledge leaves with them unless it was written down. | Runbooks, architecture notes and IP are yours at handover, by design. | Depends entirely on the contract. |
| Ongoing change | Immediate — they are already on payroll. | Retained arrangement, or hand it to your own developer. Both are fine. | Change requests, usually priced. |
| Where this fails | Expensive if the work is not continuous. | Wrong for you if you need certification, daily on-site presence, or a bench of four starting Monday. | Overhead you fund whether or not you use it. |
The honest summary: hire in-house when the work is permanent and continuous. Use a larger agency when you need scale or a compliance function. Use us when you need a specific system built properly, handed over, and off your cost base when it is done.
If you would rather talk than type, both calendars below are live and show real availability in your own time zone. No charge, no deck, and nobody will chase you afterwards.
Bring the process that costs you most. By the end you should know whether there is a project here, roughly what shape it takes, and a cost range for work of a similar shape — including when the honest answer is that you do not need us.
See available timesOften enough to work out whether there is a project here at all. If there is not, you will be told so — that answer is cheaper for both of us early than late.
See available timesPrefer to write first? The enquiry form is below, and WhatsApp is the fastest route if you want an answer the same day. Hours are Monday to Friday, 09:00–18:30 IST.
Answered here so you do not have to ask.
You describe the process and the constraints. We tell you the shape of engagement it needs and roughly what that costs, including when the honest answer is that you do not need us. There is no deck.
Yes. Pointing automation at a broken process just makes the mess arrive faster. If the fix is a process change or a cheaper off-the-shelf tool, that is what we will say — including when it costs us the work.
Usually not. Much of the value is in connecting systems you already pay for. We recommend replacing a tool only when keeping it costs more than moving off it.
You do, on final payment. Pre-existing components and open-source dependencies are listed with their licences at handover, so you know what you own outright and what you hold under licence.
Before the call: nothing to prepare. No brief, no budget figure, no shortlist. If you can describe a day in your operation, that is enough to work from.
Delivery runs from both. Remote-friendly and asynchronous by default, so which desk an engineer sits at rarely affects a project.
Our Chennai base, inside the CIT innovation park. Most SkillTank cohorts and a large share of day-to-day delivery run from here.
The registered address of KaizenSpark Tech Private Limited, and the address to use for contracts, invoices and vendor onboarding.
Visiting? Tell us which office suits you and we will confirm a time. Both are working offices rather than reception desks, so a heads-up matters.
Usually with questions rather than a brochure. WhatsApp is the fastest route if you want an answer the same day.
Spent on your process, not on our credentials. By the end you should know whether there is a project here.
Either a discovery proposal with a defined price and duration, or a note explaining why we think you should not spend the money yet.
Over fifty companies served since 2024. Most of that work sits under NDA, so what follows describes the build rather than naming the client. Named case studies get published here as they are cleared.
Five build types that recur across our client work. If yours resembles one of these, we have made the mistakes already.
Bulk CV intake parsed into structured candidate records — skills, experience, contact details — with confidence scoring and a review queue for anything the model was unsure about. Replaces the manual reading pass that stands between an application and a shortlist.
An operations console covering catalogue, orders, inventory and fulfilment, with role-based access and an audit trail. Built to take the daily running of the store out of a spreadsheet and off the founder's laptop.
End-to-end handling of an internship intake: applications, allocation, progress tracking, certificate generation and reporting — built on Google Apps Script and Workspace so it ran on infrastructure the organisation already owned.
Status updates, reminders and first-line support delivered on WhatsApp Business API, with escalation rules routing anything uncertain to a human rather than guessing. Built for a market where WhatsApp is the channel people actually answer.
A reporting and administration layer over a learning platform — cohorts, attendance, progress and outcomes — reading live operational records rather than a nightly export, so the number on screen matches the number in the system.
Many more under NDA. Ask for a walkthrough →
Modular ERP covering people, scheduling, finance and reporting. Adopted module by module rather than in one cutover.
Over 5,000 people trained through supervised work inside our own codebases, with review and pairing as the teaching method.
Standard on our engagements, not an upsell.
Under NDA by default. We do not publish client names, screenshots or numbers without written approval — which is why this page describes builds rather than naming them.
If one of the builds above resembles your problem, we can walk you through the architecture and the trade-offs on a call.
We quote per engagement after discovery rather than from a rate card. An estimate given before the process is mapped is wrong in both directions — usually expensively so. This page explains the models, what drives the number, and what is included regardless of size.
Most clients start with discovery and move into a build. Programme and retained arrangements follow from there.
The problem is clear, the solution is not.
Scope is agreed and ready to ship.
Live systems needing support and iteration.
Programme engagements — several linked systems delivered across three to six months — are priced as a sequence of build phases with a roadmap agreed up front. Agency partnerships run on agreed rates with a named engineering contact. Both are quoted individually.
If you want a cheaper engagement, these are the levers. We would rather tell you where they are than pad an estimate.
Before you speak to anyone — us included — the payback calculator works out what a manual process costs you over a year and whether a build earns that back. It returns do not build this when the running cost would exceed the saving.
How many systems the build has to talk to, and whether they have documented APIs. One legacy system without an API can cost more than three modern ones.
If the workflow is already documented and agreed internally, discovery shortens considerably. If three people describe it three ways, it does not.
Each distinct role adds permissions, screens and test paths. Two roles is not twice one — it is closer to three times.
Audit trails, data residency, retention rules and access reviews are all buildable, and all add scope. Tell us early rather than at UAT.
Moving history from an old system is frequently the largest single line item, and the one most often left out of a first estimate.
The cheapest projects have one named decision-maker who replies within a day. This is genuinely the largest variable we see.
These are part of delivery on every engagement, at every size. If another supplier quotes them as extras, that is worth comparing.
| Item | How we handle it |
|---|---|
| Currency | Invoiced in INR for clients in India; other currencies quoted on request |
| Payment schedule | Staged against delivery milestones rather than calendar dates |
| Taxes | GST applied as applicable and shown separately on every invoice |
| Change requests | Quoted and approved in writing before work starts — never absorbed silently or billed as a surprise |
| Third-party costs | Hosting, licences and API usage billed at cost, or paid directly by you on your own accounts |
| Contract | Statement of work per engagement; we work under your MSA and NDA where you have them |
On the first call we will give you a range. Not a number we then defend for six weeks — a range for work of a similar shape, with the assumptions behind it stated out loud so you can tell us which ones are wrong.
The cost of leaving a process alone is the number most people never put a figure on. Put your own numbers in — nothing is sent anywhere, and there is no email gate.
Assumes automation removes about 70% of the hours — the rest is judgement, exceptions and the cases software should not touch. It is a prompt for a conversation, not a quote.
Takes the figures straight to the enquiry form below, with the assumptions attached. No charge, and nobody will chase you afterwards.
Describe the process and the constraints. We will tell you the shape of engagement it needs and roughly what that costs — including when the answer is that you should not spend the money yet.
Answered here rather than in a questionnaire three weeks into a deal. Where a control is aspirational rather than in place, this page says so — a supplier who overstates their posture is a risk in itself.
Security handled as a delivery practice rather than a pre-launch audit that finds everything too late to fix cheaply.
Credentials held in a secrets manager or the platform's own store, never in source control, configuration files or a shared document. Rotation procedure documented at handover.
All changes reviewed before merge regardless of author or urgency. Review is the primary control against both defects and unreviewed access changes.
Dependencies pinned, audited for known vulnerabilities, and updated on a schedule rather than only when something breaks. Retained clients get this continuously.
Role-based permissions and audit trails on record changes are treated as build requirements in any system holding operational or personal data.
Development, staging and production kept separate with distinct credentials. Production access is limited and logged.
Every release has a rollback path established before deployment. Backups configured and restore-tested rather than assumed.
AI in a business process is a reliability question before it is a capability question.
Applies to retained clients. During a build, the delivery team is available through the working week and reachable on the agreed channel.
| Severity | Definition | Target response |
|---|---|---|
| S1 — Critical | Production down, or data at risk. No workaround. | Within working hours, immediately on notification; outside them, as agreed in the SOW |
| S2 — Major | Core function unusable for multiple users. Workaround painful or absent. | Same working day |
| S3 — Minor | Function impaired with a viable workaround. | Next working day |
| S4 — Request | Change request, question or enhancement. | Scheduled into the change allowance |
Stated plainly: specific response times and any out-of-hours cover are agreed per contract and written into the statement of work. We would rather commit to a number in your agreement than publish one here that does not match what you have signed.
If your procurement process has a security or vendor assessment form, send it early. We would rather fail it in week one than in week nine.
Working notes on the integrations operations actually run on — WhatsApp Business API, Google Workspace, Razorpay — and on how we scope, measure and hand over. Written by the engineers doing the work, not a content team.
An estimate produced before anyone has watched the work happen is a guess with a budget attached. What we do in the first two weeks, and why it makes the second month cheaper.
Read the piece →
Conversation pricing, template approval, the difference between marketing and utility categories, and the line items most quotes leave out.
In progress
A large share of operations run on Sheets and Drive. Treated as a real platform, with version control and error handling, Apps Script carries more than people expect.
In progress
Settlement timing, fee splits, refunds and chargebacks reconciled against your books automatically, with exceptions surfaced rather than buried.
In progress
Automation multiplies whatever it is pointed at. Three situations where we tell clients to fix the process first and quote for less work.
In progress
Architecture notes, environment setup, runbooks and the rotation procedure for secrets. The test is whether an engineer who was not on the project can operate it without a meeting.
In progressPublishing schedule: pieces go up as the work that prompted them finishes, rather than to a content calendar. If a topic here would help a decision you are making now, ask us and we will send the notes early.
No newsletter software yet. Email us and we will add you to the list we send new pieces to.
An estimate produced before anyone has watched the work happen is a guess with a budget attached. Here is what we do in the first two weeks instead, and why it makes the second month cheaper.
Most software quotes are produced from a conversation. Someone describes a problem, a supplier recognises a shape they have built before, and a number comes back within the week. It feels efficient. It is the single most expensive habit in this industry.
When we sit with the people doing a job, the work almost never matches the description we were given. A finance lead will describe invoice approval as three steps. Watching it, there are eleven — and four of them exist only because a system two departments away exports a file in the wrong format on Thursdays.
That gap is not anyone being careless. Descriptions are summaries, and summaries drop exactly the detail that determines cost: the exceptions, the re-keying, the handoffs where work waits.
A supplier who quotes from a description is pricing the summary. You will pay for the eleven steps either way — the only question is whether that happens in the estimate or in the change requests.
Two reasons, both boring. First, the integration surface stops being a surprise — we already know which systems have real APIs and which will need a scraper or a manual bridge, and that is the largest single variable in any estimate.
Second, scope arguments happen before anyone has written code. A change request in week two of discovery costs a conversation. The same change in week nine of a build costs a fortnight.
It happens, and it should. Sometimes the honest output is that a process change, a different licence tier, or simply agreeing who owns a decision removes the problem for nothing. We would rather write that down and lose the build than deliver something that automates a bad process faster.
That is not generosity. Fixing a broken process before automating it is how the automation ends up worth keeping.
Discovery is the cheapest stage of the engagement and the one that decides whether the rest is worth doing.
Pick an industry you already understand. Get 14 production-ready products white-labelled to your brand, keep 60% of every subscription plus 100% of implementation fees — and build a client book you own, can sell, and we will buy back.
See the client, the pricing and what you keep per client.
GST filing, digital invoicing, inventory control and CRM have moved from optional to mandatory. But the vendors building that software sell generic products over the phone and lose interest the moment a client needs someone who understands their trade. That gap is the entire business.
A transporter buys from someone who knows what a lorry receipt is. Whoever can run a demo in the client's own language of work wins the deal, almost regardless of price.
Fourteen production-grade products means years of engineering, hosting, security and support before a single invoice is raised.
Commission on someone else's platform ends when the deal closes. Without the client relationship and a recurring share, you are not building an asset.
You do the part that needs a human who understands the trade. We do the part that needs an engineering team.
You pick an industry, learn it deeply, run demonstrations, close the sale, collect the implementation fee and own the client relationship. You can sell anywhere in India, to any business in your vertical.
KaizenSpark engineering handles product development, cloud hosting, security patching, releases and second-line technical escalation. Your brand sits on the front; our team sits behind it.
An implementation fee you keep in full, plus 60% of a subscription that renews every month. Cash in the near term, a compounding book of recurring income over time.
A software book does not obey the same rules as a shop, a dealership or a trading business.
Compliance and digital invoicing requirements push adoption regardless of sentiment or season.
Each client adds to your base rather than resetting it. Month twelve starts from month eleven, not zero.
Nothing to stock, move, insure or write off. Working capital stays available instead of sitting on shelves.
Engineering, hosting and product development sit with us. Your cost base is largely your own selling time.
You get the full white-labelled catalogue from day one and may sell any of it. Certification in one vertical is what earns you Specialist status, priority lead routing and a higher share — because depth closes deals that breadth cannot.
Products released while you are an active partner are added to your catalogue at no additional cost.
No cities, no maps, no allocation queue. Tiers are earned on performance, and every partner enters at the same door.
Everything you need to sign your first clients.
Unlocked at 12 active clients plus vertical certification.
For firms that support and enable other partners.
Most channel programs let you earn commission. This one lets you build something you own, can hand on, and can sell. That distinction is the whole point.
Client agreements sit in your business name, under your brand, on your domain. We are your supplier, not the counterparty to your customer. Your book appears on your balance sheet, not ours.
You may sell your book to another certified partner, or pass it to a family member or successor. Our consent is required so the clients stay properly served, and it will not be unreasonably withheld.
After 24 months, if you want out for any reason, KaizenSpark will repurchase your book at fifteen times its monthly net revenue. You are never trapped in a business you no longer want to run.
Every figure below is net — your share after ours, less the platform fee. Adjust the assumptions to match the industry you know.
Charged for configuration, data migration, staff training and go-live support. We take nothing from this line.
Billed monthly on user count and data volume. Rises to 70% once you reach Specialist tier.
Illustrative model only. Not a projection, guarantee or representation of earnings. Actual results depend on your sales effort, market conditions, pricing and client retention. Figures are net of our revenue share and platform fee but gross of your own costs and applicable taxes. Year-one recurring assumes clients are signed at the start of the period.
Indicative costs of assembling the same capability independently in the Indian market — not a discount claim.
| Component | Indicative standalone cost |
|---|---|
| Development of 14 production-grade products | + |
| Cloud infrastructure and hosting, year one | |
| Ongoing engineering, patching and L2 support | |
| Brand, white-label configuration and domain setup | |
| Sales playbook, pricing sheets and demo collateral | |
| Marketing asset library and campaign creatives | |
| Certification and vertical training programme | |
| Indicative cost to replicate independently | |
| KaizenSpark activation, total |
Covers certification, white-label setup and your partner console. This is all you pay until the business is actually running.
The balance falls due only once you have proven the model works for you. If you never reach three clients, you never pay it.
| Factor | Build your own | Plain reselling | KaizenSpark Partner |
|---|---|---|---|
| Capital to start | + and 2–3 years | Low, but so is the return | one-time |
| Who owns the client | You — if you survive the build | The vendor, not you | You, under your own brand |
| Income shape | Nothing for years, then recurring | One-time commission per deal | Implementation fee plus 60% recurring |
| Product breadth | One product, if you are lucky | Whatever the vendor sells | 14 products across 14 industries |
| Who fixes bugs at 11pm | Your engineering team | A queue you cannot escalate | Our L2 team, under your brand |
| Growth ceiling | Your capital and hiring | Vendor's commission table | Tier upgrades and overrides |
This is the method our own engineering team uses on every project, and it is what we certify you in. It is also the reason partners win deals against cheaper products: nobody else walks in and maps the client's process before quoting.
Sit with the people doing the job. Trace where hours go, where data is re-keyed, which spreadsheet everyone actually trusts.
Write down what will be delivered, what it must connect to, and what is explicitly out of scope, before anything is configured.
Put something usable in front of the client in short cycles, so feedback arrives while it is still cheap to act on.
Agree the numbers up front: time per task, error rate, hours returned. Without a baseline there is nothing honest to compare to.
Kaizen means continuous improvement. Keep reading the numbers with the client and keep tightening the system. This is what makes them stay.
Four stages. Both sides confirm fit before any commitment, and the full agreement is shared before any payment.
You tell us your background and the vertical you want. We confirm whether the current cohort has room.
Thirty minutes on your vertical, your existing network, the agreement terms and mutual fit.
Agreement signed, platform white-labelled to your brand, console provisioned, vertical certification completed.
Your first client demonstrations run jointly with our team, then handed over as you close independently.
We take a limited number of partners into each cohort, because that is how many our certification and support team can properly onboard.
No, and that is deliberate. Territory models cap your growth at the size of one city and leave you blocked if a neighbouring partner underperforms. Instead you specialise by industry and can sell anywhere in India. What protects you is deal registration: register a prospect and no other partner can be supported on that account for 90 days, extending to 120 days at Specialist tier.
Deal registration, enforced through the partner console. First to register a qualified prospect holds it. If a registration lapses without progress it returns to the pool. This is the standard mechanism across enterprise software channels, and it protects effort rather than geography.
No. Hosting, development, integrations and second-line technical support are handled by KaizenSpark engineering. Your role is commercial: finding prospects, running demonstrations and owning client relationships in an industry you understand.
You pay the smaller entry amount to start. The balance of the activation fee is due only after your third client signs, and the monthly platform fee is waived until then too. If you never reach three clients, you never pay either. You keep 100% of implementation fees and 60% of every subscription, rising to 70% at Specialist tier. The complete schedule is in the agreement you review before paying anything.
You do. Client agreements are in your business name, under your brand and domain. We are your supplier, not the counterparty to your customer. That is what makes your book an asset you can transfer or sell rather than a commission stream that stops when you do.
After 24 months you can sell your book to another certified partner, pass it to a successor, or ask us to repurchase it at fifteen times its monthly net revenue. We can offer that because a repurchased book pays us the full subscription rather than our 40% share — roughly nine months of that revenue covers the repurchase.
They are never left stranded. We either support them directly or transfer them to another certified partner, at no cost to them. Your clients bought on the strength of your brand, and we will not let that be worth nothing to them.
Yes, with conditions we can verify. Complete certification, log 25 documented client demonstrations within six months, sign nobody, and your activation fee is refunded in full. The bar is deliberately about effort you control, not outcomes you do not. If a committed person follows our playbook and still cannot get to a first sale, the problem is our enablement.
It covers hosting, security patching, releases and L2 support that run whether or not you sell that month, and it lets us keep the activation fee low. It is also why we can accept partners at this price instead of charging several lakh upfront.
A full-stack IT and AI engineering company registered in Tamil Nadu since November 2024, building enterprise software, SaaS platforms and AI-driven automation across manufacturing, healthcare, education, retail, real estate, finance, professional services and logistics. We build our own products on the same stack we license to partners.
Nothing is taken away from you. Your existing clients, brand and income continue exactly as they are. Priority lead routing pauses until you are back on track, and your tier progression stalls. Because there is no territory involved, an underperforming partner blocks nobody.
Many partners start part-time and transition once recurring income is established. There is no office or staffing requirement, though consistent weekly selling time is necessary during the first two quarters.
A program that only works when partners succeed should carry consequences for us when they do not. These are written into the agreement.
Complete certification, log 25 documented demonstrations in your first six months and still sign no clients, and we refund your activation fee in full. If our training and collateral cannot get a committed person to a first sale, that is our failure, not yours.
If you ever stop trading, no client of yours is left stranded. We take over support directly or transfer them to another certified partner, at no cost to them. Your clients trusted your brand; we will not let that be worth nothing.
The revenue split, platform fee, buy-back multiple and every other charge are stated on this page and fixed in the agreement. No undisclosed deductions, no rates that change once you have signed.
Apply free to be assessed for the current cohort, or book a direct consultation to move faster.
We respond within one working day with your cohort status. No payment required at this stage.
Your details are used solely to assess this application.
For candidates ready to review the agreement in detail.
Adjusted against your entry payment on completion.
We keep each cohort small, because that is how many partners we can properly certify and support at once. Start for and pay the balance only when it works.
This page covers one of five partnership routes. If you landed here looking for something else, these are the ways in.
Some people want to sell business software under their own brand. Some simply want to introduce a client and get on with their day. Some need engineering capacity behind an agency they already run. Those are different arrangements with different obligations, so this page separates them rather than pretending one programme fits everybody.
Effort and return rise together. None of these require you to write code. The ones that pay most require you to hold the client relationship and stay in it after the sale.
You know someone with the problem. That is the whole job.
You sell and serve under your own brand. We build, host and support behind you.
You have the clients and the design. You need engineers who will not embarrass you.
Your product and ours are better sold together than apart.
Institutions and employers who want people trained on real systems.
Not sure which one you are? Most people arrive thinking they want a reseller agreement and discover a referral arrangement suits their week better — or the reverse. Describe what you do and who you already talk to, and we will tell you which route fits, including when the answer is none of them.
Whatever the route, these hold. They are the terms we would want if we were on your side of the table.
Nothing begins on a handshake and a hope. What each side owes, how money moves and how either party leaves is agreed in writing before the first client is touched.
We sign yours. If you do not have one we will provide a plain version, but the default assumption is that the document protecting your clients is the one you already trust.
A client you introduced or registered is yours. We will not approach them behind you, and we will not quietly take the renewal. If that ever happens, you have a written agreement to point at.
Every partner gets a named contact who knows the account. You are not filing tickets into a general inbox and waiting to find out who owns the answer.
We do not ask you to stop working with anyone else, and we do not carve up territories. What protects your effort is registration of the specific deal you are working, not a line on a map.
Every arrangement states how it ends and what happens to clients, data and revenue when it does. A partnership you cannot leave cleanly was never a partnership.
Four stages. Most partners are through the first three inside a fortnight.
Thirty minutes. What you sell today, who you already have access to, and how much of your week you can genuinely give this. We will tell you which route fits and, when the numbers do not work for you, we will say so rather than sign you up to find out slowly.
The specific arrangement, the money, the obligations on both sides, how deals are registered and how either party exits. You read it without a countdown timer attached. Questions get answered in writing so there is a record.
Access to what you need to actually sell: demo environments, the current product portfolio, pricing you can quote from, and the answers to the objections you will hear in week one. For delivery partners this stage is a technical walkthrough instead.
We sit in on the first one. You lead the relationship, we cover the technical ground, and afterwards we review what worked. Nobody learns a new offer from a document alone.
We would rather lose an application than take on a partner who was told what they wanted to hear.
Enough detail that we can give you a real answer rather than a brochure. We reply within one working day.
How much does this really earn. How long before it earns it. What happens when a client is unhappy. Ask those before you fill in a form — they are the questions we would ask.
You probably do. The accountant re-keying the same invoice into three systems. The distributor running on a spreadsheet nobody dares touch. Introduce them, and we take it from there — scoping, quoting, building and supporting — while you stay in the loop and get paid for the introduction.
No agreement to sign, no portal to learn, no quota. You make one introduction and we do the rest.
An email, a WhatsApp message, or the form at the bottom of this page. Tell us who they are and what is going wrong for them. Please only send someone's contact details when they are happy to hear from us — a name and a company is enough to start if they are not ready for that yet.
A short conversation to find out whether software is actually the answer. Sometimes it is not, and we will say so to them and to you. We would rather return an introduction than sell a system that does not fix the problem.
Every introduction is different in size and in how much work you did to make it happen, so the fee is agreed with you per referral rather than fixed by a published rate. It is confirmed in writing before the client signs anything, so you are never guessing at the end.
Won, lost or parked, you hear the outcome. If it converts, the fee is paid after the client's first payment clears. If it does not, you find out why, which is usually more useful than the fee would have been.
You do not need to diagnose anything. If a business you know shows one of these, it is worth sending over.
An order arrives by email, gets typed into a spreadsheet, then typed again into accounting software. Every re-keying is an hour lost and an error waiting.
What worked at eight people is breaking at twenty-five. Nobody has time to fix it because everybody is busy doing the thing that needs fixing.
They bought software, it was implemented badly, and now half the team quietly works around it. These are difficult, honest projects and we are glad to take them.
Orders, approvals and complaints all arrive in one thread. It works until somebody scrolls past a message worth a hundred thousand rupees.
The owner asks a simple question and somebody spends two days in spreadsheets to answer it, by which point the answer has changed.
There is a single person who knows how everything works, and the business stops when they take leave. That is a risk, not a process.
What we cannot use: a scraped list, a name without context, or somebody who has not agreed to be contacted. Cold lists waste your credibility and ours, and sharing personal contact details without consent creates a problem for you, not just for us.
You are lending us your credibility. These are the rules that protect it.
Two minutes. We come back to you within two working days with whether it is worth pursuing and what your fee would be.
Referral is the lightest of five ways to work with us. If you want to sell under your own brand, or bring us in behind your agency, those routes are set out in full.
Status, milestones, invoices, documents and support requests for every system we run for you — without waiting for someone to reply to an email.
Milestones, what shipped, what is in review and what is blocked - without waiting for a Monday email.
Scope, architecture notes and the handover pack for every system we have delivered to you.
Raise an issue against a specific system, watch it move, and see what was changed to close it.
The portal is a separate, secured application. Sign-in happens there rather than on this page, so your password is only ever typed into the system that can actually check it.
Request portal access →The portal is being rolled out client by client. Tell us which systems you are running and we will set your access up and send you the link.
Forgotten your password? Your account manager can reset it — get in touch and we will issue a new one.
Need an account for a colleague? Ask us rather than sharing a login. Access is granted per person, so the audit trail stays meaningful.
The portal is not a status page we remember to update. It reads the same records we work from, so what you see is what we see. More on how we work.
Twenty-one of them, answered the way we answer them on a call — including the ones where the answer is no, we do not hold that certification, or no, you probably should not build this. If your question is not here, ask it and we will add it.
We give you a range on the first call. Not a number we then defend for six weeks — a range for work of a similar shape, with the assumptions written down so you can see what would move it. Most clients start with a discovery engagement and move into a build; programme and retained arrangements follow from there. The pricing page sets out the four shapes and what sits inside each.
Because the same feature can differ by a factor of three depending on what it has to connect to. A build that talks to one documented API is not the same job as one that talks to a legacy system without one, even if the screens look identical. Publishing a single number would mean padding every estimate to cover the worst case. We would rather show you the levers and let you pull them.
Six things, in roughly this order of impact. How many systems the build has to talk to, and whether they have documented APIs. Whether the workflow is already documented and agreed internally — if three people describe it three ways, discovery takes longer. The number of distinct user roles, because two roles is closer to three times the work than twice. Compliance requirements such as audit trails and data residency, which are all buildable and all add scope. Data migration, which is frequently the largest single line item and the one most often left out of a first estimate. And decision speed on your side: the cheapest projects have one named decision-maker who replies within a day. That last one is genuinely the largest variable we see.
Testing, code review, documentation, environment setup and handover material are part of delivery on every engagement, at every size — not line items. If a competing quote lists them separately, that is worth comparing carefully, because it usually means the base number is not what it appears to be.
Usually not. A large share of the value is in connecting systems you already pay for — Google Workspace, your CRM, Razorpay, WhatsApp. We recommend replacing a tool only when keeping it costs more than moving off it. Replacing working software is expensive, disruptive, and frequently solves a problem the business did not have.
You describe the process and the constraints. We tell you the shape of engagement it needs and roughly what that costs, including when the honest answer is that you do not need us. There is no deck, and nobody will chase you afterwards.
Because a description is a summary, and summaries drop exactly the detail that determines cost. A finance lead will describe invoice approval as three steps. Watching it, there are eleven, and four of them exist only because a system two departments away exports a file in the wrong format on Thursdays. A supplier who quotes from a description is pricing the summary — you pay for the eleven steps either way, the only question is whether that happens in the estimate or in the change requests. We wrote this up in full: map the process before you quote it.
Yes, and it is often the better arrangement. We can specify and hand off, pair on delivery, or take a defined slice with a clear interface while your team keeps the rest.
Yes. Automation is a multiplier, and multipliers do not care about the sign of what they multiply — pointing one at a broken process makes the mess arrive faster. If the fix is a process change or a cheaper off-the-shelf tool, that is what we will say, including when it costs us the work.
Then the design was wrong. Anything customer-facing ships with confidence thresholds, human review on uncertain cases, and a logged trail so a failure is diagnosable rather than mysterious. Our governance page sets out the review thresholds, traceability and fallback behaviour in detail.
You do. Data stays in your accounts wherever the architecture allows, access is scoped to what each system needs, and we document data handling in writing before a build starts.
No. We hold no formal security certification at present and will not imply otherwise. What we offer instead is documented practice — secrets management, code review on everything, dependency hygiene, environment separation and reversible releases — all set out on the governance page.
Yes, and we will answer it honestly, including the questions where the answer is no. Send it to officials@kaizensparktech.com.
It is listed, scoped and disclosed before anything ships. Which components send data to a model provider, what they send, and what the fallback does when the model is unavailable are all documented on the governance page rather than left for you to discover later.
You do. Ownership of work produced under the engagement transfers to you on final payment. Pre-existing components and open-source dependencies are listed with their licences at handover, so you know exactly what you own outright and what you hold under licence.
Yes. We work under your MSA, NDA and supplier terms where you have them, rather than insisting on our paper. Each engagement has a statement of work of its own.
KaizenSpark Tech Private Limited, incorporated November 2024, registered in Tamil Nadu, India. Registration details are available on request for vendor onboarding — ask and we will send them.
Because we do not publish client names, screenshots or figures without written approval, and most of our clients would rather we did not. That is why the work page describes builds rather than naming them. We would rather be the supplier who protects a client’s confidence than the one with the better-looking website. If you want references, ask on the call and we will arrange them properly.
Architecture notes, environment setup, runbooks, the rotation procedure for secrets, and the documentation needed to operate the system. The test we apply is whether an engineer who was not on the project can run it from the documents without a meeting. Documentation is written during the build rather than after it, which is the only way it ends up accurate.
They are agreed per contract and written into the statement of work. During a build the delivery team is available through the working week and reachable on the agreed channel; response commitments and any out-of-hours cover are set out in your agreement. We would rather commit to a number you can hold us to than publish a general promise that means nothing.
You can. Documentation, runbooks and environment setup are delivered as part of the build specifically so you are not dependent on our availability. A supplier who makes leaving difficult is managing your dependency rather than your system.
Describe the process and the constraints. We will tell you the shape of engagement it needs and roughly what that costs — including when the honest answer is that you do not need us.
