The price per document is fixed, the number of documents is not
Digital access looks like a rate negotiation and it is not. The unit price is banded and largely predetermined by volume, while the volume itself is produced by a measurement whose assumptions are contestable. Buyers who negotiate the rate and accept the count are optimising the smaller of the two variables.
Prepared by Redress Compliance · August 11, 2026 · SAP advisory. Based on 30 to 40 SAP indirect access reviews, 2024 and 2025.
Executive summary
SAP measurement tools overcounted by 20 to 45 percent before duplicate and follow on records were filtered out of the line item base.
The tooling counts what it can see rather than what the rules charge for, and the difference is not a rounding error, it is a quarter to nearly half of the proposed bill.
That single fact determines where the buyer side effort should go, because a 30 percent overcount at any rate is worth more than a hard fought improvement on the rate itself.
Sales and invoice documents drove 60 to 80 percent of total chargeable volume in order driven businesses. Both carry full cost weight, which means the exposure concentrates in the two document types with no weighting relief available.
Any estate with ecommerce, EDI, or bot driven order entry should expect its number to be dominated by these two, and should scope its verification effort accordingly rather than spreading it evenly across all nine types.
Buyers who scoped the measurement before SAP ran its own count paid 30 to 55 percent less than buyers who accepted the vendor run baseline. That gap is not a discount, it is the cost of conceding the definition of the countable population.
Once a vendor produced number is on the table it becomes the reference point that concessions are measured against, and every subsequent argument is framed as a request for relief rather than a correction of fact.
Only the initial creating document counts, and follow on records inside the same business process do not. This rule is where most of the overcount lives, because a sales order that becomes a delivery and then an invoice looks like three documents to a naive count and is one chargeable event.
Buyers who cannot demonstrate process linkage in their own data end up paying for the same transaction more than once.
What counts, and at what weight
| Document type | Typical source | Cost weight |
|---|---|---|
| Sales | Ecommerce, CRM, EDI | Full |
| Invoice | Billing engines, portals | Full |
| Purchase | Procurement, supplier portals | Full |
| Service and maintenance | Field service tools | Full |
| Manufacturing, quality, time, material | MES, IoT, HR feeds | Reduced or exempt |
The weighting table is the reason a document count and a chargeable count are different numbers.
Four of the nine types carry full weight and the rest carry a reduced multiplier or fall outside scope entirely, so an estate with heavy operational telemetry can present a large raw count and a modest bill, while an order driven estate with a smaller raw count can present a much larger one.
The practical consequence is that the composition of the count matters as much as its size, and any measurement that reports a single headline figure without the type breakdown cannot be evaluated. Ask for the breakdown first, because a number you cannot decompose is a number you cannot argue with.
The wider model sits in the SAP licensing guide.
Scoping the measurement before the vendor does
- Establish process linkage in your own data, because only the initial creating document is chargeable and follow on records inside the same process are not, which is where most of the 20 to 45 percent overcount originates.
- Isolate sales and invoice volume first, since these two full weight types drove 60 to 80 percent of chargeable volume in order driven businesses and no weighting relief is available on either.
- Separate the reduced weight and exempt types explicitly, so that operational telemetry from manufacturing, quality, time, and material feeds is not carried into the bill at full value.
- Identify duplicate records before the vendor tool runs, because a duplicate in the line item base is indistinguishable from a genuine document once a vendor produced count has been published.
- Produce your own defensible number before the measurement lands, which is what separated the buyers who paid 30 to 55 percent less from those who negotiated down from a vendor baseline.
The SAP digital access playbook
The measurement scoping method, the document type breakdown, and the top ten negotiation recommendations.
Get the playbook →The rate is banded, the count is argued
It is worth separating the two variables in this pricing model, because buyers routinely spend their effort on the one with less movement in it.
The per document rate starts near fifteen cents in the entry band and steps down as committed volume rises, and while there is negotiation available in which band you land and how the tiering is structured, the shape of that curve is broadly standardised.
The count is a different kind of number entirely.
It is produced by a measurement whose result depends on how duplicates are treated, whether process linkage is established, and how the reduced weight types are separated from the full weight ones.
And our reviews found that measurement running 20 to 45 percent high before those filters were applied.
A 30 percent overcount at any rate costs more than a hard fought improvement in the rate at any count, and the two are not equally contestable. The rate is a commercial position that SAP defends deliberately.
The count is a factual claim that has to survive inspection, and inspection is something the buyer can do in advance. This is why the 30 to 55 percent gap between scoped and unscoped buyers exists, and it is not a discount gap.
Once a vendor produced number is on the table it becomes the anchor, and every correction the buyer makes afterwards is heard as a request for relief rather than as a factual dispute about what the rules actually charge for.
The remedy is sequencing rather than skill: build your own defensible count first, present it with the type breakdown and the process linkage behind it, and let the vendor measurement be the thing that has to reconcile to yours.
The audit posture around this sits in the SAP license audit survival guide.
- Percentile standing for your exact deal size and industry, from real closed transactions
- Scenario simulation before the call: test alternative terms and see the financial impact of each
- A negotiation playbook, talking points, and a two page executive brief on day one
What we saw across SAP digital access engagements, 2024 and 2025
Across roughly 30 to 40 SAP indirect access reviews run between 2024 and 2025, the first document count SAP proposed ran well above the count the buyer could defend, and the pattern was consistent enough to plan against:
How far SAP measurement tools ran above the defensible count before duplicate and follow on records were filtered out of the line item base.
The gap between buyers who produced their own defensible count in advance and those who negotiated down from the vendor run baseline.
Three patterns recurred: measurement tools overcounting by 20 to 45 percent before duplicate and follow on records were filtered, sales and invoice documents driving 60 to 80 percent of chargeable volume in order driven businesses.
And buyers who scoped the measurement first paying 30 to 55 percent less than those who accepted the vendor baseline.
The buyer side move is to build a defensible count with its type breakdown and process linkage before the measurement lands. The wider library sits in the SAP practice.
Your first five moves
- Build your own document count before SAP runs one, because the buyers who did paid 30 to 55 percent less, and the saving comes from owning the anchor rather than arguing with it.
- Establish process linkage in the line item base, since only the initial creating document is chargeable and unlinked follow on records are the largest single source of the 20 to 45 percent overcount.
- Verify the sales and invoice volume line by line, as these two full weight types carry 60 to 80 percent of the bill in order driven estates and no weighting relief applies.
- Demand the type breakdown with any count, because a single headline figure cannot be evaluated and the composition drives the bill as much as the size does.
- Negotiate the band placement alongside the count, then hold both. The SAP practice runs the measurement and the negotiation together.
Frequently asked questions
How does SAP price indirect access in 2026?
On documents created in the digital core by non SAP systems, not on the named users behind them. The metric is volume based and charged once per initial creating document. List price starts near fifteen cents per document in the entry band and steps down as committed volume rises.
Which variable actually moves the bill, the rate or the count?
The count. The rate curve is broadly standardised and SAP defends it deliberately, while the count is a factual claim produced by a measurement that ran 20 to 45 percent high in our reviews before duplicates and follow on records were filtered out.
A 30 percent overcount costs more than a hard fought rate improvement.
Why do the measurement tools overcount?
Because they count what they can see rather than what the rules charge for. Only the initial creating document in a business process is chargeable, so a sales order that becomes a delivery and then an invoice is one chargeable event that a naive count reads as three.
Duplicate records in the line item base compound it.
Where does the exposure concentrate?
In sales and invoice documents, which drove 60 to 80 percent of chargeable volume in order driven businesses. Both carry full cost weight and no weighting relief is available on either, so verification effort should be concentrated there rather than spread evenly across all nine document types.
Do all nine document types cost the same?
No. Sales, invoice, purchase, service and maintenance carry full weight. Manufacturing, quality, time, and material documents carry a reduced multiplier or fall outside scope.
This is why the composition of a count matters as much as its size, and why a headline figure without a type breakdown cannot be evaluated.
How much does scoping the measurement first actually save?
Buyers who produced their own defensible count before SAP ran one paid 30 to 55 percent less than buyers who accepted the vendor run baseline. That is not a discount gap.
It is the cost of conceding the anchor, after which every correction is heard as a request for relief rather than a factual dispute.
What about the Digital Access Adoption Program?
It offered a transition discount of up to 90 percent for measured conversion, which makes the measurement even more consequential rather than less.
A large discount applied to an inflated count still produces an inflated bill, and the discount is a one time event while the count sets the recurring baseline.
Digital Access Travels With You
Session 7 of the SAP RISE Migration Series. Nine document types, roughly one FUE per thousand documents, and the most expensive misconception in the RISE conversation: more than half of buyers assumed the move removed indirect access exposure. The first document volume becomes the renewal floor.
