A modernization strategy, ready in three weeks
We establish what the platform does, where the risks sit and which route fits each part of it. Fixed price, and the result stays with the organization.
The first conversation is free of charge.
- Runtime
- 3 weeks
- Investment
- €5,000 – €15,000, fixed
- Result
- owned by the organization
The platform still runs, the next decision is unclear.
That is the situation the assessment is made for.
Plan a first callSeven possible outcomes, chosen per part of the platform.
Different parts often get different routes. Keep is a real outcome, and that advice is part of the service.
Keep
- What it means
- The system stays as it is
- When it fits
- It does its job, and change costs more than it gains
Which routes are likely in play?
Get an initial assessment in 1 minute.
How well does the system support the work today?
The routes the assessment would look at first
Based on these answers, these routes are the most likely candidates.
- KeepLikelyThe system stays as it is
- RehostLikelyThe system moves to new infrastructure
- ReplatformLikelyParts of the stack change, the logic stays
- RefactorLikelyThe code is cleaned up, the behaviour stays the same
- RearchitectLikelyThe structure changes, the logic is kept
- RebuildLikelyThe system is rewritten from the requirements
- ReplaceLikelyA standard product takes over the function
The answers point in different directions. That usually means different parts of the platform need different routes, which is exactly what the assessment sorts out.
One answer was "not sure", so this indication is rough. Finding out is part of the first week of the assessment.
An indication, not advice. The assessment decides the route per part of the platform, and sometimes the advice is to change less than expected.
Four lines of inquiry, one documented picture.
- Before week 1
Scope, price and read access agreed.
- Week 1
Interviews with users, management and engineering. Documentation set against what the system really does.
01InterviewsUsers, management and engineering. The people who work around the system every day know where it fails, and that knowledge is rarely on paper.
02Documentation and functionalityWhat the system does, set against what is documented. The gap between the two is usually where the risk sits.
- Week 2
Semi-automated code review: AI reads the codebase, an engineer checks everything. Infrastructure and architecture.
03Semi-automated code reviewAI helps read the existing codebase and reconstruct thin documentation, so years of business logic surface within weeks. An engineer checks everything.
04Infrastructure and architectureWhat the platform runs on, what is still supported, and what a modern foundation would need to carry.
- Week 3
A route, phasing and cost per part. Readout to the decision makers.
Routes, phases and costsEach part of the platform gets a route, with its pros and cons, a phasing and a cost estimate per phase.
ReadoutWe present the findings and our recommendation to the people who take the decision.
A report a director can sign, a plan an engineer can use.
Written so any competent team can act on it, including one other than ours.
Read access and a few conversations, the operation keeps running.
- Read access to code and history
- Documentation, however thin
- Interviews with users, management and engineering
- One contact person
- Detailed solution design
- Implementation
- Penetration test
An ERP nobody could replace, and the route that left it standing.
Unsupported since 2020, running the production line, twenty years of data.
- 01No upgrade path
- 02Data access was the problem, collection was not
- 03Data quality first, then AI
- 04The first step for AI across the group
- 1Replace the ERPConsidered
Heavy investment and a production stop.
- 2Buy SaaS and connect itConsidered
Still dependent on the old data.
- 3Build an accessible data layerChosen
Reaches the data without touching the ERP.
- 4Do nothingConsidered
Data stays out of reach.
A fixed price, agreed before we start.
Fixed before we start, and confirmed in the first conversation.
- Runtime
- 3 weeks
- Result
- stays with the organization
- First conversation
- free
Questions about the assessment
Straight answers to the questions that come up before an assessment starts.
How long does the assessment take?
Three weeks, counted from the moment access to the code and the documentation is arranged.
What does it cost?
Between €5,000 and €15,000, depending on the scope. The price is fixed before we start.
Is the first conversation free?
Yes. The first conversation about the assessment is free of charge and without obligation. The assessment itself is paid.
What if the advice is to change nothing?
Then that is the outcome. Sometimes the advice is to change less than expected, and that advice is part of the service.
Does the organization have to build with Eli5 afterwards?
No. The organization keeps the result whether or not we build anything afterwards, and the report is written so another team can act on it.
How is AI used in the assessment?
AI helps read the existing codebase and reconstruct thin documentation. An engineer reviews everything. The models run on European hosting, in line with the GDPR and the AI Act.
What if the only developer who knows the system leaves?
That is precisely the risk the assessment maps. The code review reconstructs documentation from the existing codebase, so the knowledge no longer sits in one head.
How much time does the organization need to invest?
Interviews with users, management and engineering, spread over the first two weeks, plus one contact person and the readout.
Every route starts with one conversation.
The first call is with a partner, free of charge.
