Analytics Legends The knowledge platform for SAP Analytics
Guide

SAP BW modernisation assessment — what a real one produces, and what it costs

As of 2026-08-14

A SAP BW modernisation assessment exists to make one expensive decision safe: which migration path this estate takes, and what the execution phase will therefore cost. It is a bounded engagement — in our engagement data, discovery work of this kind lands between €50k and €150k — whose entire job is to enable a decision on an execution phase that commonly runs from €500k into the millions. An assessment that produces a slide deck of options without ruling any out has not done that job, whatever it cost.

The four deliverables that make an assessment worth buying

First, an object inventory with complexity scoring. Every BW object gets a score, and the distribution of those scores — not the object count — is what sizes the programme. On a Tier-1 estate, seventy to eighty per cent of objects typically fall in the auto-eligible band and twenty to thirty per cent require manual redesign; the assessment's job is to tell you where this estate sits against that shape, because the manual tail is where the money goes.

Second, a cohort assignment with its consequence stated. Classic BW under extended maintenance, ECC alongside BW, already on BW/4HANA, or departed to an open stack — each has a different trigger, timeline and risk appetite, and the assignment changes what the programme is even trying to buy.

Third, a shortlist of migration paths with at least one explicitly ruled out and the evidence for ruling it out. An assessment that keeps every option alive has deferred the decision it was paid to close.

Fourth, the customer's own maintenance terms, quoted from their documents rather than asserted from a schedule someone half-remembers.

What the complexity distribution actually tells you

Scores of one and two generate automatically with under five per cent manual touch-up. A three still generates, but expect twenty to forty per cent manual remediation on it. Fours and fives are flagged for manual redesign, and pushing those through automated generation anyway is a common and costly mistake: the output runs, but reconciliation against the legacy report fails late — during user acceptance testing rather than during build, which is the most expensive place to discover it.

This is why an assessment that reports "1,400 objects" without a score distribution has told you almost nothing. Two estates with identical object counts and different complexity tails produce programmes that differ by a factor that no amount of negotiation recovers.

The same logic applies to what does not convert: custom ABAP transformations, complex CompositeProvider unions and process chains needing orchestration outside the generator's scope. Those surface as conversion warnings, and counting them is a deliverable, not a footnote.

The test that separates an assessment from a sales artefact

Ask what the assessment has ruled out and on what evidence. The single most damaging conflict of interest in this market is recommending a full rebuild when the underlying model is salvageable, because rebuild engagements are larger — experienced buyers have started screening for it explicitly, and firms seen doing it lose exactly the long-term relationship the migration was supposed to create.

The mirror-image failure is recommending a lift-and-shift because it is quick to sell, without naming the debt that travels with it. Both failures look identical on the deliverable: a recommendation with no path visibly rejected.

A second test: ask who owns report retirement. Selective migration is viable when a named business sponsor owns retirement decisions and stalls when nobody does. An assessment that has not asked this question has not found the programme's actual long pole, which is political rather than technical.

Where the demand for these assessments actually sits

Read the live market rather than the pitch. On our own opportunity radar, BW appears in roughly a quarter of live SAP analytics engagements at the time of writing, and Germany alone accounts for close to half of those BW-mentioning briefs — ahead of India, Poland, the United States, Spain, Austria, France and the United Kingdom.

That concentration matters when you are choosing who to buy an assessment from. A firm whose BW migration references are all in one country is not disqualified, but the pattern tells you where its delivery bench actually is, and a Tier-1 assessment is executed by the bench rather than the logo.

The wider commercial frame is well established: the migration cycle SAP triggered by setting a maintenance horizon on BW and naming Business Data Cloud as the analytics target is the largest in this ecosystem since BW moved onto HANA, and the associated consulting demand is measured in billions through 2028. That is the market the assessment is a first step into, and it is why the discovery phase is priced as a decision instrument rather than a report.

The decision the assessment has to close

The model is in reasonable shape and the driver is infrastructure or licensing pressure → lift-and-shift, framed explicitly as transitional, with a debt register attached.

The estate is large, reporting cannot be frozen, and a multi-year modernisation programme is funded → selective migration, with a sunset date recorded in the architecture decision at kickoff.

The estate is small or well documented → direct remodelling in the target platform is faster and cleaner than any bridge.

The model itself is the problem — undocumented ABAP exits, a decade of layered workarounds, reporting definitions finance has never trusted → rebuild, because every other path rehosts the defect faithfully.

The estate works and the real gap is reach rather than replacement → stay and federate. It is a legitimate and under-sold answer, and an assessment that never surfaces it is not neutral.

Nobody owns report retirement → none of the above will hold its schedule. Fix sponsorship before scoping, because that is the finding that saves the most money and is the least often written down.

What we cannot assert

SAP does not publish a price for a modernisation assessment, and we do not publish a formula for one. The bands on this page are our own engagement observations on Tier-1 estates; they are not quotes, and they do not survive a change of estate size or country. The radar shares describe the live opportunity feed on the date this page was generated and move as the feed moves.

Frequently asked

How long should a BW modernisation assessment take?

It is a bounded discovery engagement, not an open-ended study: its purpose is to close a path decision so the execution phase can be scoped and funded. If it has not ruled a path out, extending it rarely helps — the missing input is usually sponsorship, not analysis.

What does a BW modernisation assessment cost?

Discovery work of this kind lands between €50k and €150k in our engagement data, against execution phases that commonly start around €500k. Those are our own figures, not published SAP pricing, and they describe Tier-1 estates.

Can the assessment be done with tooling alone?

Tooling produces the object inventory and complexity scores; it does not produce the cohort assignment, the ruled-out path or the retirement sponsor. The mechanical share is generated, the residue is judgement, and the residue is where assessments earn their fee.

Does the maintenance extension make the assessment less urgent?

It removes the compliance framing — SAP moved the BW 7.5 extended-maintenance window to 31 December 2030 from 2027 — but not the economics. Tooling advantage, delivery capacity and sequencing against any S/4HANA programme all argue for the same window.

What this page is built on

Read next