Negotiating AI add-ons with your ERP vendor at renewal
how to negotiate AI add-ons with your ERP vendor
Also searched as
- ERP vendor AI licence negotiation
- ERP renewal AI copilot bundled pricing
- ERP vendor AI contract data rights clause
Short answer
Treat the AI add-on as a separate decision from the renewal, and ask for four things in writing: unbundled pricing with consumption caps, no training on your data, full API and data access regardless of the AI SKU, and clean exit terms. Use a credible alternative as leverage, not as a threat.
Applies to: Renewals and expansions with ERP vendors and resellers (Infor, Oracle, SAP, Microsoft, Epicor and others). Not legal advice; involve procurement and counsel.
A negotiation sequence that works
- 1Separate the AI add-on from the base renewal in the proposal. Ask for each as its own line with its own term, so a discount on one is not paid for by the other.
- 2Ask what is included in the base subscription today. Many vendors are putting assistants into existing tiers; do not buy as an add-on what is becoming standard.
- 3Request the pricing metric in writing: per user, per tenant, per consumption unit, per credit. Ask what happens when credits run out and whether overage is capped.
- 4Get data terms: your data and prompts are not used to train shared models, retention periods for prompts and outputs, sub-processors, and data location.
- 5Secure API and data access terms that do not depend on buying the AI SKU, so a different assistant or reporting layer can read your data through supported interfaces.
- 6Negotiate term and exit: short initial term or pilot period, termination for convenience on the AI line, price caps on renewal, and data export on exit.
- 7Run a time-boxed evaluation of the vendor's assistant against an alternative on the same questions from your data, and share the comparison method, not your numbers.
- 8Route the draft to legal and security. Confirm every verbal promise appears in the order form or amendment.
Where the leverage actually is
Vendors want AI attach rates, and early adopters are valuable references. That gives you room on price, pilot terms and caps in exchange for a clear evaluation and a credible decision date. It does not give you leverage if you are locked into a multi-year commitment with no decision coming; know your renewal calendar well in advance, ideally a year before expiry.
Your best leverage is an operational alternative. A private assistant running on open-weight models against read-only ERP access shows the vendor that the AI layer is replaceable. You do not have to deploy it fully. A bounded pilot is enough to make the alternative real, and it informs the comparison honestly.
Be fair about the trade-off. The vendor's assistant has deep product knowledge, supported updates and no integration work. For in-product tasks it may be the better choice. Negotiate for terms that let you use it where it is strong without blocking other tools elsewhere.
Clauses to read closely
Data and training: look for language on whether your content improves vendor or third-party models, how long prompts are retained, who can review them, and whether sub-processors such as model providers are named. For export-controlled or defense-related data, confirm whether processing location and personnel meet your own obligations; your compliance lead decides, not the sales engineer.
Consumption: credits, tokens or 'AI units' can be opaque. Ask for a worked example of a typical query's cost on your data, a usage dashboard, alerts at thresholds, and a hard cap option. Ask whether unused credits roll over.
Access and portability: ensure bulk data export, supported APIs and event interfaces are not gated behind the AI tier. Check auto-renewal and price uplift terms, and whether features you rely on can be moved to a higher-priced SKU later. Ask for a most-favoured-customer or price cap clause on renewal where possible.
Questions to send before the renewal meeting 1. Which AI features are in our current tier today? 2. Pricing metric and overage terms in writing? 3. Is customer data used to train any model? Retention of prompts? 4. Named sub-processors and processing locations? 5. Are APIs/data export available without the AI SKU? 6. Termination for convenience on the AI line? Price cap at renewal? 7. Can we pilot for 90 days with a conversion option?
Making it a decision, not a drift
The common failure is accepting a bundled discount that quietly raises the base price for several years while the AI usage stays low. Compare three-year total cost for the base alone, base plus AI, and base with a separate private layer. Make the comparison on the same scope and the same users.
Document the decision criteria before the vendor sees them: accuracy on your question set, data controls, cost per active user, integration effort, and exit cost. Score both options on the same sheet. That record also helps your board and auditors later.
Finally, keep relationships constructive. You still need the vendor for upgrades, support and roadmap. A firm, specific, polite ask for unbundled and portable terms is usually received better than a hostile one.
Common pitfalls
- !Signing a multi-year AI commitment before a pilot proves usage.
- !Accepting a discount on the AI line that is funded by a higher base uplift.
- !Verbal assurances on data use that never reach the order form.
- !Ignoring credit or token mechanics until the first overage invoice arrives.
- !Letting API and export rights depend on the AI tier, which creates lock-in.
- !Starting negotiation inside the last 60 days before renewal, when leverage is weakest.
How an ERP-grounded AI assistant handles this
A grounded assistant such as ERPray can support the negotiation prep from your own side: it can report how many named users actually transact in each module, which features and screens are used, and which customisations would break if a module changed. That gives you evidence for right-sizing the renewal and for deciding where a vendor assistant would really be used. It does not interpret contracts for you; legal does.
Frequently asked questions
Should we buy the vendor's ERP AI assistant at renewal?
Only if a pilot on your own data shows it answers your real questions well enough, on terms you can exit. For in-product tasks it may be a good fit and simplest to run. Avoid a long bundled commitment before you have usage evidence, and keep API access independent of the AI tier.
What data-use terms should we insist on?
Written confirmation that your data and prompts are not used to train shared models, defined retention for prompts and outputs, named sub-processors, stated processing locations, and breach notification terms. Have counsel and security review them, especially if you handle export-controlled or customer-restricted data.
How do consumption credits usually go wrong?
Credits are consumed faster than expected when users run broad questions or agents loop over large datasets, and overage can be priced above the committed rate. Ask for a worked cost example, usage dashboards, threshold alerts and a hard cap so a busy month cannot become a surprise invoice.
Is it realistic to threaten to leave for a private AI layer?
Threats are weaker than demonstrations. A bounded pilot of a private assistant that reads the ERP through supported read-only access makes the alternative credible without disrupting the ERP itself. Present it as an evaluation, set a decision date, and keep the conversation about terms, not ultimatums.
Related
Does Your ERP Vendor's AI Copilot Increase Lock-In?
Native vendor copilots such as SAP Joule, Oracle Fusion AI agents and Microsoft Copilot in Dynamics 365 are usually bundled or low-cost to start, but they only see that vendor's native data model and typically bill on consumption, which raises both switching cost and long-run cloud spend. Use them for what they do well inside the vendor's own UI, and keep cross-system or custom-field questions on a vendor-neutral layer.
CIO briefingERP AI Vendor Due Diligence: What CIOs Should Actually Check
The single highest-signal question in ERP AI vendor due diligence is 'show me it answering a real question against our actual ERP data, live, not a demo dataset.' Beyond that, focus diligence on data grounding method, security and access model, exit/portability terms, and references from customers on your specific ERP platform - not on feature checklists, which most vendors can match on paper.
CIO briefingERP AI budget line items: what to put on the sheet before finance asks
Budget ERP AI in eight groups: infrastructure, software and model licences, integration, data preparation, security and compliance, people and change, run cost, and contingency plus exit. The costs that surprise teams are data cleanup, internal time and ongoing evaluation, not the model itself.
CIO briefingERP AI: when to hire a partner versus build it in-house
Build in-house when you have or can hire engineers with real experience in retrieval-augmented generation, your ERP's specific API/data model, and production LLM deployment, and when the AI capability is a genuine long-term differentiator worth owning. Hire a partner when time-to-value matters, your team's strength is running the ERP rather than building AI infrastructure, or you want to validate the use case with a working pilot before committing to a multi-quarter internal build. Most mid-size manufacturers are better served starting with a partner-led pilot and deciding on in-house investment once the use case is proven.
CIO briefingERP Upgrade or AI First? A CIO Decision Framework
In most cases, add a grounded AI layer on top of your current ERP first, because it proves value in weeks and shows exactly what an upgrade would need to fix. Reserve a full ERP replacement for cases where the platform itself is end of support, unsupported, or structurally blocking the business, not simply because it feels dated.
CIO briefingHow Much Does ERP AI Actually Cost?
A scoped pilot against one ERP module typically runs 15,000 to 50,000 USD; a single-department production deployment runs 50,000 to 150,000; a multi-module enterprise rollout runs 150,000 to 500,000 or more, plus ongoing hosting and usage costs. Native vendor copilots are usually priced per active user per month on top of existing licensing, not included free.
AI for ERPAI for Infor SyteLine and CloudSuite Industrial
Add grounded AI to Infor SyteLine or CloudSuite Industrial: natural-language answers, agents over IDOs and ION, on-prem or CloudSuite deployment.
AI for ERPAI Agents Over Infor ION API, BODs, and the Infor Data Lake
Build AI agents on Infor ION API Gateway, BODs, and Data Lake, alongside or instead of Infor's own GenAI features, with your choice of model and hosting.
Stuck on ERP AI (any ERP)?
Talk to engineers who work inside ERP AI (any ERP) every week, and who build private AI that answers these questions from your own ERP data.