.png)
Frameworks, core principles and top case studies for SaaS pricing, learnt and refined over 28+ years of SaaS-monetization experience.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Procurement Guide How Are Omnichannel Order Management Systems OMS Priced for Enterprise Retailers
Enterprise retailers rarely buy an order management system because the current platform lacks another dashboard. They buy because inventory is fragmented, orders are being cancelled, stores cannot fulfil digital demand, returns move slowly, or fulfilment decisions are eroding margin. Yet the commercial proposal often measures something only loosely connected to those problems: order lines, processed orders, gross merchandise value, users, API capacity, locations, modules, or a negotiated cloud commitment.
The choice of meter matters more than the opening discount. A retailer processing three million orders a year can pay almost the same under order-based and order-line pricing when baskets are small. At four lines per order, however, the order-line bill can be more than three times as high, even when the number of customers served has not changed. Meanwhile, quote-only platforms can hide five-year commitments, implementation services, API packs and support charges behind one attractive annual subscription.
Monetizely’s position is clear: enterprise retailers should make fulfilled customer orders the primary OMS pricing meter, supported by a fixed platform fee, declining volume rates, a seasonal burst allowance and a hard overage cap. Per-user pricing does not track OMS value, while uncapped order-line pricing makes product assortment and basket size a vendor revenue lever.
The meter, not the feature list, determines who carries growth risk
Monetizely’s 5-Step Pricing Framework begins with Goals and Segmentation, which identifies the commercial goal and the customers being served. Packaging decides which capabilities sit together and which are sold separately. Pricing Metric selects the unit that makes the bill rise as customer value rises. Rate Setting determines the actual price, discounts and volume curve. Operationalisation turns the design into working entitlements, usage records, invoices, renewals and revenue recognition. The sequence matters because an attractive metric cannot repair confused packaging, and a sound rate cannot survive poor metering. As discussed in Monetizing Agentic AI, operationalisation often demands more work than the visible price design because billing, entitlement and finance systems must all agree on what has been consumed.
For OMS procurement, packaging, metric and operationalisation carry most of the risk. Retailers usually know why they need the platform. They struggle to establish exactly which orders count, which capabilities require add-ons, how peak volume is measured and what happens when an acquisition, marketplace launch or store rollout changes usage.
The scorecard below grades six leading enterprise approaches against those three steps. The grades assess commercial clarity and buyer alignment, not product quality.
| Vendor and offer | Packaging | Pricing metric | Operationalisation |
|---|---|---|---|
| IBM Sterling Order Management | B+ - Four editions create a visible progression, but store engagement, call centre, resilience and other capabilities use separate units. | C - Order lines track workload, yet they charge more for larger baskets even when customer-order value is unchanged. | B- - Public units help benchmarking, but SaaS, container, RVU, user, device and instance measures complicate the order form. |
| SAP Order Management for sourcing and availability | B - The core sourcing and availability offer is comparatively focused. Wider order-to-cash needs can pull buyers into other SAP products. | A- - A processed sales order is close to the value event a retailer recognises. | B - Annual blocks and a clear transaction definition help, although auto-renewal and block depletion need contract controls. |
| Salesforce Order Management | B+ - Order Visibility, Growth, Commerce bundles and Order Management Lite support several entry points. The same range also creates overlap. | C- - Per-processed-order and GMV allocations coexist, while orders entering outside the Commerce storefront can attract separate charges. | C - Managed and unmanaged order rules, API allocations, locations, support plans and bundled limits require sustained licence administration. |
| Manhattan Active Order Management | B - A broad unified-commerce suite supports expansion, but public module-level pricing is absent. | D+ - Commercial rates are not public, while technical entitlements visibly scale by subscribed modules and order throughput. | C+ - Long cloud terms offer continuity, but buyers need explicit rights for API capacity, modules and implementation services. |
| Fluent Order Management | B - A composable offer can limit forced suite adoption and is available through Google Cloud Marketplace. | D - No public rate card or primary commercial meter was disclosed as at 4 August 2026. | C+ - Marketplace procurement can simplify billing, but it does not by itself make order definitions, overages or renewal rates transparent. |
| Oracle Fusion Cloud Order Management | C+ - The product covers order capture, orchestration, pricing and fulfilment, but often sits inside a wider Fusion SCM programme. | D - Oracle’s public product pages did not disclose an OMS rate or usage meter as at 4 August 2026. | C- - User access, suite dependencies, environments, implementation partners and extensions can all expand the effective scope. |
IBM and SAP show that public usage pricing is feasible. Salesforce shows the risk of combining several value measures. Manhattan, Fluent and Oracle leave more of the commercial design to the negotiation, which makes buyer discipline more important than list-price comparison.
OMS pricing falls into three broad contract shapes. The first charges for operational throughput, usually processed orders or order lines. The second prices a broader commerce platform through GMV, allocations or editions. The third uses a negotiated enterprise subscription whose internal volume assumptions may appear only in the order form.
Public terms available on 4 August 2026 make the split visible.
The headline rate is only the first layer. IBM’s public table also lists add-ons measured by authorised users per month, client devices per year, instances per month, customer value units and annual order lines. Salesforce Premier support is published at 30% of net licence fees. Manhattan’s technical documentation states that Unified Commerce API capacity varies by subscribed module and orders per hour, with extra volume packs available for sustained higher usage.
Those details explain why two retailers buying the same OMS can receive very different commercial outcomes. One may purchase orchestration alone. Another may require store fulfilment, call-centre tooling, inventory availability, reverse logistics, premium support, extra environments and integration capacity.
Vendor economics also favour longer commitments. Manhattan generated USD 408.1 million in cloud-subscription revenue in 2025, up 21% from 2024, while its cloud agreements typically ran for at least five years. SAP reported €21.9 billion of current cloud backlog for the first quarter of 2026. Christian Klein described SAP as having a “strong start to the year” as customers expanded across its suite.
Retail buyers should expect sales teams to trade discount for duration and scope. A 25% discount on a five-year commitment can still cost more than a three-year contract that gives the buyer rights to reduce capacity, remove modules and rebaseline after a divestiture.
Order-line pricing turns basket complexity into a tax
A processed customer order is a visible commercial event. The retailer accepts one checkout, makes one promise and has one customer relationship to protect. An order line is a technical component inside that event.
The difference becomes material for grocery, marketplaces, department stores, beauty, home improvement and B2B retail, where one checkout can contain many products. Bundles, partial fulfilment, substitutions and returns can create further lines without producing another customer order.
IBM’s August 2026 public Standard price of USD 0.045 per line per month equates to USD 0.54 per annual order line. SAP’s published price equates to USD 0.657 per processed sales order per year. Those rates appear similar only when orders contain roughly one line.
The following model holds order growth constant and changes basket complexity. It shows what a retailer could pay over three years before discounts.
| Modelled contract | Three-year volume | Published or modelled software spend | Modelled three-year TCO |
|---|---|---|---|
| SAP processed-order pricing | 10.42 million orders | USD 6.84 million | USD 12.84 million to USD 16.84 million |
| IBM Standard with 1.2 lines per order | 12.50 million order lines | USD 6.75 million | USD 12.75 million to USD 16.75 million |
| IBM Standard with 2.4 lines per order | 25.00 million order lines | USD 13.50 million | USD 19.50 million to USD 23.50 million |
| IBM Standard with 4 lines per order | 41.67 million order lines | USD 22.50 million | USD 28.50 million to USD 32.50 million |
| Quote-only platform subscription | USD 2.5 million in year one with 5% annual increase | USD 7.88 million | USD 13.88 million to USD 17.88 million |
The model demonstrates the flaw in treating an order line as a neutral measure. At 1.2 lines per order, the IBM and SAP public rates produce almost identical software spend. At four lines, the IBM line-based amount reaches USD 22.5 million, compared with USD 6.84 million under SAP’s processed-order measure.
A line-based vendor may argue that more lines require more computation. That claim has some merit, but cloud cost does not usually rise in a straight line with the retail price charged. Many OMS activities happen at order, fulfilment or event level. Inventory checks may occur before any order exists, while one order can generate repeated sourcing decisions, status changes and API events.
Fluent Commerce has noted that availability checks can outnumber orders by ratios ranging from 2:1 to 700:1, depending on how and where availability is displayed. Manhattan’s 2026 API documentation likewise allocates Unified Commerce capacity according to modules and orders per hour, not simply annual order lines.
Those engineering facts strengthen the case for a two-part price. A fixed platform fee pays for persistent availability, security, environments and baseline capacity. A per-order charge captures growth in realised activity. Technical limits can govern abusive API patterns without turning every system event into a billing event.
Quote-only suites make contract mechanics the real price
Opaque pricing does not automatically mean expensive pricing. Large retailers can negotiate substantial discounts from Manhattan, Salesforce, Fluent or Oracle. The problem lies in comparability: without a public rate or stable unit, the buyer cannot tell whether the quote is low because the vendor is discounting, because important scope is excluded, or because the commitment is larger than the initial forecast.
Manhattan’s disclosures show why services deserve particular attention. Professional services produced USD 503 million, or 47% of total 2025 revenue, compared with USD 408.1 million from cloud subscriptions. Cloud software may be the strategic product, but implementation, planning, consulting and training remain economically significant.
Eddie Capel, then Manhattan’s president and chief executive, said in October 2024 that the company’s “fundamentals are strong” and highlighted top-line growth, profitability and continued innovation. Those strengths are attractive to buyers, but they also give a leading vendor room to hold firm on term, scope and services.
Salesforce creates a different form of complexity. Commerce Cloud charges a percentage of storefront GMV, while orders processed through external channels but using Salesforce Order Management can generate a separate OMS charge beyond included limits. Order Management licences can also carry managed-order, unmanaged-order, GMV, location, data-storage and API allocations.
A retailer with Salesforce Commerce, Service Cloud and OMS therefore needs a channel map before signing:
- Orders placed on the Salesforce storefront must be separated from marketplace, call-centre, social-commerce, subscription and third-party storefront orders.
- Managed orders must be distinguished from orders held mainly for visibility.
- Returns, exchanges, test orders, fraud cancellations and failed imports need written counting rules.
- Included API, storage, inventory-location and support allocations must be modelled against peak demand, not annual averages.
The following decision matrix shows which commercial structure best serves common retail segments.
A well-negotiated metric follows the retailer’s operating model without making the contract unique beyond auditability. The buyer should not accept a formula so bespoke that only the vendor can calculate it.
Public OMS-specific pricing disputes are scarce because enterprise order forms, statements of work and settlement terms are generally confidential. Public filings and government records from adjacent enterprise programmes provide a clear warning: the worst cost shocks have come from access definitions, customisation, failed implementation and weak governance, not merely from the listed licence rate.
Four documented cases show the scale.
These were not OMS licence-overage cases, and they should not be presented as such. They involved wider SAP, IBM and Oracle programmes. The procurement lesson remains direct because an OMS connects commerce, payments, inventory, warehouse systems, carriers, stores, tax engines, customer service and finance. A weak integration or data model can stop orders even when the software itself remains available.
Queensland Premier Campbell Newman said in August 2013, “It appears that IBM took the state of Queensland for a ride.” The state barred IBM from new contracts until the company improved governance and contracting practices.
Birmingham finance lead Brigid Jones took a less adversarial tone in June 2023 but reached a similar governance point: “there are lessons to be learned from what has happened.” The council’s report said a move from adopting standard processes to adapting Oracle around existing processes had severely affected implementation.
The pattern should change how an OMS RFP is evaluated. A bidder offering a USD 2 million subscription and USD 8 million implementation is not cheaper than a USD 3 million subscription and USD 4 million implementation. Nor is the first offer safer when the services estimate excludes data repair, carrier onboarding, store rollout, peak testing or post-launch support.
Operationalisation also includes commercial data. The retailer needs a daily usage feed showing counted orders, excluded transactions, order lines, source channels and overage position. Finance should be able to reproduce the invoice without asking the vendor to explain its own calculation.
The best 2026 deal names processed orders as the primary meter
Monetizely’s preferred 2026 architecture begins with a fixed annual platform fee. That fee covers the production tenant, agreed non-production environments, security, updates, baseline availability and a defined level of technical capacity.
The primary variable meter should be fulfilled customer orders. An order should count once when it reaches an agreed state, such as released for fulfilment or accepted by a fulfilment node. Failed imports, test traffic, duplicate messages, fraud rejections and vendor-caused reprocessing should not count.
Order lines can remain a guardrail rather than the invoice driver. For example, the base rate could include an average of three lines per order across the contract year. Only a sustained rise beyond that band would trigger a prospective rate review. Such a design protects the vendor from a radical workload change without making every extra item an immediate charge.
The rate curve should fall as volume grows. A retailer that doubles orders usually does not double the vendor’s sales, onboarding, account-management or platform costs. Declining unit rates share those economies and keep the software affordable as digital penetration rises.
A credible enterprise offer would therefore contain:
- A fixed platform fee that represents no more than 25% to 40% of expected recurring software spend.
- A committed annual order band with declining unit rates at higher volumes.
- A seasonal burst pool allowing agreed peak throughput without overage.
- A maximum annual charge, normally 110% to 120% of the committed subscription, unless the buyer acquires a business or adds a new brand.
- Enterprise pooling across countries, banners and channels rather than separate minimums.
- Prospective true-ups and credits for unused commitments, not retroactive list-price overages.
Before signature, buyers should also complete the following negotiation checklist.
SAP currently comes closest to the preferred core meter because it defines one transaction as one processed sales order. IBM gets important elements right by publishing editions, indicative prices and units. Its weakness is making the line the main economic event. Salesforce gets packaging breadth right, but the overlap between GMV, processed-order allocations, external-channel charges and support percentages can make the full bill hard to forecast.
Manhattan gets long-term platform continuity and unified capability right. Its next commercial step should be a standard public meter definition, even if rates remain negotiated. Fluent gets composability and cloud-marketplace procurement right, but enterprise buyers still need a disclosed consumption model. Oracle gets end-to-end process breadth right, while public pricing transparency remains the central weakness.
Our committed view is not a choice among unrelated pricing models. Processed orders should be the primary meter across enterprise OMS deals. A fixed platform floor, declining order rates and tightly defined capacity rights form one coherent architecture. Seats, lines, GMV and API calls may be used as safeguards or secondary entitlements, but none should control the main bill.
Buyers preparing a 2026 procurement should act at a higher level before opening commercial negotiations:
Create an enterprise order model before issuing the RFP. Establish one reconciled view of orders, lines, channels, returns, cancellations and peak throughput using at least 24 months of data.
Choose the preferred meter in the procurement strategy, not after vendor selection. Requiring processed-order bids prevents finalists from controlling the comparison through incompatible pricing structures.
Fund the operating change separately from the software decision. Store processes, inventory accuracy, fulfilment rules and customer-service workflows need executive ownership beyond the technology programme.
Keep the systems integrator commercially independent from the licence recommendation. A partner paid more for a larger suite or more custom work should not determine the target architecture without buyer challenge.
Judge the investment against avoided cancellations, fewer split shipments and lower fulfilment cost. Licence savings alone do not justify a platform that makes poor sourcing decisions or delays order release.
Assumptions
The three-year model starts with three million orders in year one, 15% annual order growth and the public IBM and SAP rates accessible on 4 August 2026. The quote-only scenario uses a USD 2.5 million first-year subscription with a 5% annual increase. Each scenario adds USD 6 million to USD 10 million for implementation, integration, testing, internal operations and support. Figures exclude discounts, taxes, currency movements, acquired businesses and financing costs.
Footnotes
https://www.amazon.com/Monetizing-Agentic-AI-Handbook-Transformation/dp/B0H7Z13VKJ/
IBM Sterling Order Management pricing, accessed 4 August 2026: https://www.ibm.com/br-pt/products/order-management/pricing
SAP Order Management for sourcing and availability pricing, accessed 4 August 2026: https://www.sap.com/products/crm/sourcing-and-availability.html
Salesforce Order Management licences and allocations, accessed 4 August 2026: https://help.salesforce.com/s/articleView?id=commerce.omlicenses.htm&language=enUS&type=5
Salesforce Commerce Cloud pricing and external-channel order rules, accessed 4 August 2026: https://www.salesforce.com/in/commerce/b2b-ecommerce/pricing/
Manhattan Associates 2025 Form 10-K, filed February 2026: https://ir.manh.com/node/21211/html
Fluent Commerce Google Cloud Marketplace announcement, 2 October 2025: https://fluentcommerce.com/resources/news/fluent-commerce-now-available-on-google-cloud-marketplace/
Oracle Fusion Cloud Order Management product page, accessed 4 August 2026: https://www.oracle.com/es/scm/order-management/
Manhattan API rate-limit documentation, accessed 4 August 2026: https://developer.manh.com/docs/reference/api-limits/
SAP first-quarter 2026 financial results, 23 April 2026: https://www.sap.com/investors/en/financial-documents-and-events/recent-results.html
Fluent Commerce discussion of OMS scalability and availability checks: https://fluentcommerce.com/resources/blog/order-management-system-pourquoi-le-volume-des-commandes-nest-pas-un-indicateur-fiable-de-sa-scalabilite/
Manhattan Associates third-quarter 2024 results, 22 October 2024: https://ir.manh.com/node/20426
SAP UK Limited v Diageo Great Britain Limited EWHC 189: https://www.bailii.org/ew/cases/EWHC/TCC/2017/189.html
Revlon 2018 Form 10-K, filed 28 March 2019: https://www.sec.gov/Archives/edgar/data/887921/000088792119000004/rev201810-k.htm
Queensland Government Health Payroll Commission of Inquiry announcement, 13 December 2012: https://statements.qld.gov.au/statements/71195
Birmingham City Council Oracle stabilisation and optimisation plan, 19 June 2023: https://www.birmingham.gov.uk/news/article/1349/plantostabiliseandoptimisecouncilitsystemsdueforcabinet_approval
Queensland Government action concerning IBM contracting, 7 August 2013: https://statements.qld.gov.au/statements/72961