SAP lease abstraction: lease accounting, lease management and RE-FX contract data for SAP

RE-FX is organised as a hierarchy. A business entity holds buildings and land, which hold rental units, and a contract attaches to those objects. The contract itself carries the terms, and the valuation rule turns those terms into postings against FI-GL and Asset Accounting. None of it starts until the terms exist as data.

Here is what has to come out of the lease document before the contract is complete.

RE-FX has a genuinely useful feature here that is easy to miss: statistical conditions let you record costs incurred before commencement, such as broker commissions or installation, so they land in the right-of-use asset calculation without generating a payable. It only helps if somebody found those numbers in the lease first.

This is the question worth being precise about, because the honest answer is not a simple no.

SAP RE-FX is a contract management and posting module. It does not parse documents. Lease terms enter it by manual entry or by data load. Separately, SAP does offer Document Information Extraction as a service on SAP BTP, and it is a capable OCR and extraction engine. But its standard document types are invoices, purchase orders, payment advice and business cards. A commercial lease is not among them. To point it at a lease you would define a custom document type, build a schema, and train it on your own documents.

So SAP can be made to extract lease data. It is a project, not a feature, and it starts from a general-purpose engine with no knowledge of what a base year is or why the third amendment overrides the original rent table. That gap between a configurable extraction service and something that already understands lease language is the whole reason this page exists. The same pattern shows up across the enterprise platforms: we walk through it for IBM TRIRIGA lease abstraction and for Yardi as well.

Large SAP shops usually end up choosing between running leases natively in RE-FX or bolting on SAP Lease Administration by Nakisa, which was built to sit alongside SAP and carry very large portfolios. SAP PRESS published a 435 page book covering both paths (Hofmann, Lim, Mabborang and Teunissen, 2018, ISBN 978-1-4932-1689-5), which tells you something about how much configuration is involved either way.

The comparison that actually matters for a data project looks like this.

We compare the suite itself on the Nakisa alternative page. The short version: if what you are missing is a subledger, buy a subledger. If what you are missing is the lease data, no subledger will produce it for you.

Four failure modes account for most of the rework, and every one of them is a document problem rather than a configuration problem.

• Amendments abstracted in isolation. A contract keyed from the original lease carries a rent and an expiration the fourth amendment already replaced. RE-FX will happily value the wrong number. This is the single most common source of a restated schedule.
• The wrong start date. Contract start in RE-FX drives the entire valuation. If it was taken from the execution date rather than the date the space was made available, every period after it is wrong.
• Options keyed as a default. Renewal terms included or excluded without a documented reasonably-certain assessment produce a lease term nobody can defend two years later, which is exactly when the auditor asks.
• Free rent and allowances lost. A rent-free period keyed as a zero condition instead of part of the schedule, or a tenant improvement allowance that never reaches the model, both distort the right-of-use asset from day one.

Source-linked abstraction does not remove the judgement calls. It makes them visible, which is the part that survives an audit.

On a migration the lease workstream tends to run on a fixed date somebody else set. The sequence that works is boring and it is the one that finishes.

1. Inventory the document stack first. Original lease plus every amendment, side letter and estoppel, per property. Teams routinely find they are missing documents for a tenth of the portfolio, and that is much better discovered in week one than in UAT.
2. Abstract, do not key. Run the documents through abstraction and review the output against the source. Reviewing a source-linked abstract is roughly a ten minute job per lease. Reading the lease cold is two to four hours.
3. Map once, load many. Agree the mapping from abstract fields to RE-FX contract, conditions and valuation inputs on a sample of ten leases, then apply it to the rest. Remapping 300 rows by hand is where the projected saving evaporates.
4. Reconcile against what you already post. Compare abstracted rent to the cash actually leaving AP. Differences are either an abstraction error or a payment somebody has been making wrong for years. Both are worth finding.

If the volume is a one-time migration spike rather than an ongoing need, lease abstraction services covers the outsourced route, and the lease abstraction software pricing comparison shows what the major platforms actually publish. For the accounting side of the same project, the ASC 842 implementation guide walks the standard end to end.