Your SAP BW system isn't a database. It's 15,000 decisions that no one has written down.

WavePath analyzes a real (anonymized) black-and-white landscape: 15,254 objects and 4,451 transformations, all mapped in a single pass.

Every SAP customer I come across is running the same countdown clock. BW is on its way out, Datasphere is where everyone is headed, and the board has realized that “soon” isn’t a date. It all makes perfect sense. There’s just one catch, and it pops up as soon as someone actually opens the system.

Fifteen thousand objects. Thousands of transformations, written by people who left years ago for better-paying jobs. ABAP routines that “do something important,” though no one can say exactly what if you ask. DataStores that were last loaded around the time the first iPhone was launched, but still quietly feed a report that someone will fight to keep.

This is the sentence that the migration slide deck conveniently skips over: moving data is the easy 10%. The difficult 90% is archaeology: figuring out what to keep, understanding how the logic actually works, and discovering what breaks the moment you point your finger at it. It’s precisely within that 90% that a 12-month plan quietly turns into a 30-month program, complete with its own steering committee, its own mythology, and a budget rule that leaves the CFO speechless.

We built WavePath to close that 90% gap. Or, as I recently told a client with more honesty than diplomacy: “We now have a real chance to actually migrate this beast.”

First, it calls the scope into question

Most migrations start with a heroic impulse: to convert everything. That single decision is exactly what turns a one-year plan into a three-year plan. Everything is in scope, so everything is suspect, so nothing goes live.

WavePath starts from the opposite premise. It reads your live BW metadata and your actual usage and load statistics, and based on that, proposes a scope that aligns with what the business is actually doing. Defunct providers, outdated loads, and source feeds that no one has consulted since the organizational chart looked completely different are automatically eliminated. You walk into the steering committee meeting with a justifiable list, not just a wish list and a prayer.

Classic 3.x and 7.x versus BW/4 object distribution across the entire catalog, immediately visible.

Then it reads the ABAP file that no one wants to open

This is the part that traditionally takes up a consultant’s entire summer, including vacation time.

WavePath’s conversion engine translates BW transformation logic routines, master data lookups, formulas, and field operations into usable SQL for your target platform. And, refreshingly, the system knows what it doesn’t know: anything that cannot be translated with absolute certainty is flagged for human review. It never quietly guesses and sneaks it through to production. Each output file clearly shows what was automated, what needs a second look, and what requires a complete rewrite.

This way, you’ll get off to a flying start with a clear to-do list. No more black box that you have to double-check line by line at midnight, while wondering whether it’s the machine or your career that’s at stake.

Each transformation is rated based on complexity and convertibility: automatic, guided, or manual.

It finds the landmines before you step on them

The most costly moment in any migration is discovering a problem after you’ve already built three layers on top of it.

WavePath reverses that order. Before a single object is rolled out, it provides a readiness assessment: object-by-object decisions, roadblocks, and the specific business questions that need an owner and an answer. The result: risks end up where they belong on a slide, in a planning meeting, with a name attached instead of as an incident at 2 a.m. with your phone lighting up the bedroom.

And it gives you a figure you can stand by in a tough meeting

Dependency order in waves, with fully manual versus WavePath-accelerated processing side by side, including ingestion and replication.

Full details about dependent waves

Wavepath does not lack any transformation or update rules

And it displays the field mapping, origin, and ABAP code at a glance.

WavePath schedules the entire migration in order of dependencies, in waves, and estimates the effort in two ways: fully manual versus WavePath-accelerated, with ingestion and replication calculated separately rather than lumped together.

So when the steering committee asks the two questions that really matter “How long, and how do you know?”you’ll have an answer backed by evidence. Not just a gut feeling disguised as a Gantt chart.

From analysis to deployment, with an assistant that knows your system

Once the plan is ready, use WavePath to create iterations (yes, with AI) of the models that need to be rolled out. It knows the old situation (BW), it knows the new situation (e.g., Datasphere), and comes back with the best-looking model you’ve ever seen not counting my son, of course. Then it pushes the generated objects directly to your target platform. The built-in AI assistant lets anyone ask questions in plain language about any object, transformation, or dependency. It queries the actual, parsed metadata in real time and can deploy on your behalf.

All of this is brought to life in a single interactive dashboard lineage, waves, roadmap, sizing, usage that runs entirely offline. Your metadata never leaves your own environment.

What WavePath Supports

For anyone scanning this list looking for their specific SAP abbreviation: yes, it’s probably there.

Sizing. How many workloads, how old are they, and how much redundancy is needed? Which queries are actively used, with which calculated and restricted key metrics, and how do you harmonize them?

Query usage. Estimated migration effort report.

Using Queries

Estimated Effort for Report Migration (Edited) Restore original

Target platforms: SAP Datasphere · Snowflake · Databricks · Microsoft Fabric · ANSI SQL.

SAP Standard Data Products. A real-time mapping of classic SAP data sources to standard SAP Data Products, always incorporating the latest SAP products. This lets you see right away what you can simply replace instead of rebuilding.

Live mapping of classic SAP data sources to SAP Standard Data Products. Always includes the latest data products.

Live deployment. CSN push to Datasphere (CLI) · Snowflake · Databricks Unity Catalog · Fabric Warehouse (including Service Principal authentication for MFA tenants).

Logic conversion. ABAP → SQL (including Datasphere CSN) for transformations: routines (IF/CASE), START delete filters, master data lookups (→ LEFT JOIN), CONCATENATE / field splits / arithmetic operations. Undefined rules are flagged and are never translated incorrectly.

BW object coverage. ADSO/ODSO/CUBE, HCPR/MPRO/HYBR/ISET, TRFN/UPDR, RSDS, InfoObjects, InfoSources (TRCS), classic 3.x flows (yes, even those!), multi-JOIN and UNION ALL views.

Analysis. Complexity score · dependency lineage · usage-driven scoping · load/recency scan · aggregation of key metrics.

Planning. Topological wave planning · manual versus accelerated effort estimates · ingestion/replication sizing.

Output. Interactive dashboard · multi-sheet Excel · SQL organized by wave · readiness and validation backlog.

AI assistant. Powered by Claude, with real-time tool access across all metadata and deployment directly from the chat.

The Conclusion

WavePath doesn’t replace your migration team. It does something more useful: it provides them with the tedious, risky, summer-consuming 90% of the data—already mapped, converted, and scored. That way, their expertise goes where it really makes a difference, not just where the spreadsheet happens to say it should.

Are you planning an SAP BW → SAP Datasphere migration (or to Snowflake, Databricks, or Fabric)? I’d be happy to run your landscape through it and show you what comes out. Some things will surprise you. Most of them will probably cause you a little—or quite a bit—of concern.

In any case: better now than in production.