
What Is a VPAT and Why Do Companies Ask for It?Read More

A focused EdTech MVP usually takes three to six months from discovery to a controlled launch. A production platform built for schools, universities, or enterprise buyers takes closer to eight to twelve months, and a high-stakes assessment system can run twelve to eighteen. Those ranges only become useful once two things are clear: which launch date is being estimated, and what the project starts from. This guide shows how to place a project in the right band, what moves it, and how to plan backward from a date that is already fixed.
The EdTech platform development timeline depends first on what kind of product it is and who has to be able to use it on day one. The ranges below come from the planning model in Arbisoft's guide to EdTech software development costs, and they cover discovery through a controlled launch with a team that has built education products before.
| Project | Typical planning range | What usually sets the pace |
|---|---|---|
| Focused MVP: one audience, one core workflow, no institutional integrations | 3 to 6 months | How well discovery defines the core workflow, and how firmly the team holds scope |
| Production platform for institutions: multi-tenant, SSO, LTI, reporting, accessibility conformance | 8 to 12 months | Integrations, admin tooling, and the security and accessibility evidence buyers ask for |
| High-stakes assessment: proctoring, item banking, audit trails | 12 to 18 months | Reliability during exam windows, integrity controls, and preparing the item bank |
| Extending Open edX or Moodle | Usually shorter to a first release than a custom build | How far the product departs from the platform's built-in model |
| Modernizing or migrating a live platform | Delivered in phases | Data quality, compatibility with what users rely on today, and keeping the current product running |
"EdTech platform" covers far more than an LMS. A tutoring marketplace, an AI study companion, a certification system, an authoring tool, and a corporate academy all count, and each has a different core workflow. The overview of EdTech platform types is a good place to check which category a product actually sits in before estimating anything, because two products with similar screens can share almost none of the work underneath.
Most timeline disagreements come from people estimating different dates. An EdTech project has at least five, and they can be months apart.
| Date | What is true on that day | What often sits between it and the next date |
|---|---|---|
| Development complete | The planned features are built and tested by the team | Bug fixing under real use, infrastructure hardening |
| First usable release | A controlled group (one cohort, one school, one client) uses the product for real work | Pilot feedback, usability fixes, missing admin tools |
| Production ready | The product is stable, monitored, and supported for its intended audience | Load testing against real peaks, support processes, documentation |
| Institution ready | SSO, required integrations, accessibility documentation, and security and privacy answers are in place | Procurement, security review, data agreements |
| Full rollout | Content is migrated, staff are trained, and every intended user has access | Training schedules, migration waves, the academic or business calendar |
A vendor quoting four months and another quoting nine for the same brief may both be right. The first is often describing development complete. The second is describing institution ready. Asking each one "which of these five dates does your estimate end on?" usually explains most of the gap.
University and K-12 buyers feel the difference most. A platform can be technically finished and still wait weeks for a security review, an accessibility check, or a signed data agreement, and procurement runs on its own schedule. Those steps belong in the plan from the start, with named owners on both sides, because nobody can speed them up at the end.
Two teams with identical feature lists can face very different schedules because they start from different places. Build means several distinct things in EdTech, and each one moves the critical path somewhere else. The complete guide to EdTech software development covers how to choose between these routes; the focus here is what each one does to the schedule.
Building from scratch gives full control and the longest path to a first release, since accounts, roles, content structure, progress tracking, and admin tools all have to exist before the product's distinctive feature can be tested. It earns that time when the learning mechanic itself is what customers pay for.
Starting from an open-source foundation such as Open edX or Moodle removes the foundation work, and the time saved goes into the parts that make the product different. The trade-off arrives later, at upgrade time: customizations that fight the platform's model turn every new release into a migration project. Arbisoft's work with Philanthropy University is one example of this route. The team integrated Open edX with NodeBB for discussion and built web and mobile apps designed for low-bandwidth and intermittent connections, and the platform now has over 100,000 registered users. The hard engineering there was connectivity, and starting from an existing learning platform kept effort pointed at it.
Extending a platform the organization already runs is usually the fastest way to add a capability, as long as the existing codebase is in reasonable shape. A short technical assessment at the start is cheap insurance, because the estimate depends heavily on what that assessment finds.
Integrating specialist products (payments, video conferencing, proctoring, identity) can be faster than building them. Integration still takes real time, since each connection needs a data contract, error handling, and testing against the other vendor's environment.
Modernizing a live platform and migrating to a new one are the hardest to compress, because the current product has to keep working throughout. These projects usually ship in slices: one service, one data domain, or one user group at a time. The total timeline is longer, but users see improvements sooner and the risk of a single big cutover drops.
The critical path is the longest chain of work in which each step depends on the one before it. Adding developers shortens parallel work but does little for that chain. In EdTech, the items below are the ones that most often end up on it.
Every role adds screens, permissions, notifications, and test paths. A consumer course product might need learners and an internal admin. A B2B training product can need learners, instructors, client administrators who manage their own employees, managers who see only their team, and support staff with limited access. Teams that scope only the learner experience tend to discover the admin half midway through the build, which is the most expensive time to find it. Writing out every role and what each one must do on launch day is one of the cheapest ways to protect a timeline.
Single sign-on, LTI launches inside an institution's LMS, SIS rostering, HRIS sync, and payment providers all depend on someone else's documentation, sandbox access, and response time. The engineering is often modest compared with the waiting. Requesting test credentials and a technical contact during discovery, well before development reaches that work, removes a common source of slipped dates.
Some EdTech timelines are limited by content rather than code. A publisher turning a print curriculum into an adaptive product needs content broken into small, tagged pieces before any personalization can work. An assessment provider needs a calibrated item bank. A platform migration needs clean historical records. These workstreams need their own owner and schedule, and they can start on day one while engineering builds the foundation.
Connecting a language model takes days. Making it reliable enough for learners takes much longer: grounding answers in the product's own content, building a test set from the subject matter, setting guardrails, deciding where a human reviews output, and running a pilot to see how it behaves with real students. Most of the schedule goes into evaluation and review, which is where AI features most often get underestimated.
Live video and real-time features bring their own infrastructure and testing across devices and networks. A native mobile app adds a second codebase, app store review, and release management. Offline use adds sync and conflict handling, which is a product decision as much as an engineering one. A responsive web app or progressive web app is often enough for a first release, with native apps added once usage shows they are needed.
Institutional buyers ask for accessibility conformance documentation, privacy terms, and security answers before they sign. Building accessible components from the first design adds little time. Retrofitting accessibility into a finished interface takes far longer and tends to land right before a launch or a deal. Products for children carry extra consent and data-handling work, which needs legal input early so engineering builds the right consent flow the first time.
Education traffic arrives in spikes: term starts, assignment deadlines, exam windows, compliance deadlines. A platform has to be load tested against the realistic peak before the first one hits, and that testing belongs in the schedule before launch.
These are hypothetical projects, included to show how the drivers above change a plan.
A startup has a working consumer study app and a pilot university interested in licensing it for next fall. The consumer product is on version three and stable. The tempting estimate is "a few weeks to add a university plan."
The university, though, needs students to sign in with campus SSO, the tool to launch from inside its LMS through LTI, instructors to see their own sections only, an accessibility conformance report, and answers to a security questionnaire. Each item is modest on its own. Together they add a new role model, two integrations that depend on the university's IT team, and documentation work outside engineering. Counting back from the start of term, with time for the university's own review and a short pilot, the real deadline for development is several months earlier than the founders assumed. The useful decision is which items must be ready for the pilot cohort and which can follow before full rollout.
A professional training company sells courses to individuals through its own academy and now wants to sell seats to employers. The learner experience barely changes. The platform needs client organizations, client administrators who assign seats and see their own employees' progress, invoicing per organization, and often SSO with the employer's identity provider.
Almost all of the new work sits on the admin and data side, and the biggest early decision is whether each employer needs its own configuration and data boundary (multi-tenancy) or whether scoped roles inside one environment are enough. That one choice shapes the data model, so settling it in discovery protects the rest of the timeline.
The phase durations below come from Arbisoft's EdTech product development roadmap, for a mid-complexity MVP. Several phases overlap, which is why the total lands at three to six months rather than the sum of every row.
| Phase | Typical duration |
|---|---|
| Discovery | 2 to 4 weeks |
| UX/UI design | 3 to 6 weeks |
| Architecture planning | 2 to 4 weeks, often overlapping with design |
| MVP development | 8 to 16 weeks |
| QA | 2 to 4 weeks, plus testing throughout |
| Pilot testing | 4 to 8 weeks |
| Launch preparation | 2 to 3 weeks |
Discovery is the phase most often cut to save time and the one that most often costs time later. A two-week discovery that settles roles, integrations, and the launch definition protects the twelve weeks of development that follow.
Many EdTech projects start with a date that is already fixed: a term start, a cohort launch, a customer contract, a board commitment. Working backward from that date gives a more honest plan than estimating forward and hoping it fits.
A shorter timeline comes from narrowing the work. The approaches below save time without creating a prototype that has to be rebuilt.
Scoping the first release around one complete workflow for one primary user group works better than a thin version of every feature. A first release where learners can finish a course and administrators can run it without engineering help tests more than ten half-built features.
Starting from an existing platform for the commodity parts (accounts, enrollment, content delivery) and building only the differentiating part shortens the path to first users. Running content preparation, integration access, and design in parallel with early engineering removes waiting from the critical path. Designing the data model for the next customer type, even when the first release doesn't use it, avoids the rebuild that often follows a first enterprise deal. Releasing to one cohort or one client before everyone shrinks the blast radius of whatever the pilot uncovers.
Some shortcuts cost more time than they save. Skipping discovery, deferring accessibility, building integrations against assumptions instead of real sandbox access, and launching without load testing all tend to come back later as larger rework.
Estimates for the same brief can differ by months for reasons that have little to do with how fast each team works. The usual causes are different launch definitions, different assumptions about integrations and migration, and whether discovery, QA, pilot support, and accessibility work are inside the number at all.
A few questions make estimates comparable:
The answers say more about a partner than the number itself. The guide on how to choose an EdTech development company covers the wider set of questions worth asking before signing.
A focused EdTech MVP can reach a controlled launch in about three months when it has one primary user group, one core workflow, no institutional integrations, and a clear scope from discovery. Three months is rarely enough for a platform that institutions will buy, because SSO, LMS or SIS integrations, admin tooling, and accessibility and security evidence add work that a consumer MVP can skip.
A custom LMS with courses, enrollment, assessment, reporting, and institutional single sign-on usually falls in the production-platform range of about eight to twelve months. Building on Open edX or Moodle typically shortens the path to a first release. Multi-tenancy, LTI certification, accessibility conformance, and migrating content from an existing LMS all push the timeline longer.
AI usually adds time, mostly in evaluation and review rather than integration. Connecting a model is quick. Grounding answers in the product's content, building a subject-specific test set, adding guardrails, and piloting with real learners takes weeks to months depending on how much the AI's output affects grades, credentials, or learner safety.
Customizing Moodle or Open edX is usually faster to a first release, because accounts, courses, enrollment, and grading already exist. The advantage shrinks when the product's core workflow departs from the platform's model, since heavy customization adds work at every platform upgrade. The approach suits products whose difference lies in content, audience, or experience rather than in the learning mechanic itself.
An EdTech platform migration takes as long as the slowest of three things: cleaning and mapping the existing data, rebuilding the workflows users rely on, and moving users without disrupting a term or training cycle. Most migrations run in phases by user group or data domain, scheduled around the academic or business calendar, so the full cutover often spans several release cycles.
Pick the launch date that matters, write down which of the five dates it means, and list what must work for the first real users on that day. That one page turns a vague question into a plan a development partner can estimate accurately.
Trusted by top platforms for our transformative solutions and exceptional results:






