How do ERP vendors get on the shortlist when buyers ask AI?

Who signs off on an ERP purchase, and what is one contract worth?

A committee led by finance and IT buys it, often with an independent consultant and an implementation partner involved.

ERP touches finance, operations, supply chain and IT at once, so the decision sits with a group: typically the chief financial officer, the chief information officer, operations leaders and, for larger projects, the board. Independent selection consultants such as Panorama are often hired to run the process, and an implementation partner or systems integrator is chosen alongside the software. That makes ERP one of the few software categories where several outside advisers stand between the vendor and the buyer.

The buyer profile spans a wide range. Software Path found that the system most often being replaced in its sample was QuickBooks, meaning a growing company buying its first ERP, and that 14% of companies were moving off homegrown or outgrown systems. At the other end are global firms moving from SAP ECC. Cloud is now the default expectation: almost 97% of companies in the same data were considering a cloud-based system, and only 3% were looking exclusively for on-premise software.

What a customer is worth is high and long-lived. With $9,000 of budget per user and 26% of employees using the system on average, by that arithmetic even a mid-sized company’s project runs well into six figures, before multi-year subscription renewals. Panorama warns that the approved figure understates the real spend: “an implementation estimate is a scope document expressed in dollars,” and hidden costs surface later. For the vendor, that means one won ERP contract can be worth more than a large volume of low-intent website traffic.

Why are so many companies evaluating ERP now?

Because of maintenance deadlines, the move to cloud and the promise of AI built into the system.

SAP’s 2027 deadline forces its large on-premise base to decide: move to S/4HANA, typically through RISE with SAP, pay for extended maintenance, or reconsider vendors. SAP’s own RISE page (opens in a new tab) pitches the migration with AI-enabled assistants for custom code, data and testing, and argues that AI embedded in cloud ERP can act within live business processes. Microsoft’s comment that ERP bookings stayed healthy while CRM slowed suggests that ERP spending is holding up even when other application budgets are tight. At the smaller end, Odoo (opens in a new tab) says it has 28 million users, showing how crowded the market for growing companies has become.

Every one of these decisions starts with research, and each involves questions buyers can now put to an AI assistant before they talk to anyone.

Where do AI assistants sit in ERP research?

Early, at the long list and the education stage, with buyers checking the answers with people later.

There is no ERP-specific survey of AI use yet. The best recent evidence is general: in a Gartner survey (opens in a new tab) of 645 business buyers, 45% said they used generative AI during a recent purchase, mainly to gather information on vendors and products, and 69% preferred to validate what the AI told them with a sales representative. Buyers in the same survey worried about misleading AI answers. That pattern fits ERP well: buyers use AI to frame the category and build a long list, then lean on consultants, references and demos to check it.

A reasonable expectation is that consultants and implementation partners use assistants too, to scan options for a client in an unfamiliar industry. Because one adviser can shape several selections, being described accurately in those answers matters more in ERP than the raw number of buyers using AI suggests.

Which questions do ERP buyers ask AI assistants?

Questions about industry fit, comparisons, migration paths, cost, implementation partners and compliance.

We wrote the example prompts below to show how a CFO, CIO or selection consultant might phrase an ERP question; none comes from observed data:

• Industry fit: “Best ERP for a $150M discrete manufacturer with three plants and make-to-order production.”
• Comparison: “NetSuite vs Dynamics 365 Business Central for a distributor outgrowing QuickBooks.”
• Migration: “Should we move from SAP ECC to RISE with SAP or evaluate other vendors before 2027?”
• Cost: “What does a cloud ERP implementation cost for 60 users, including partner fees?”
• Partners: “Which partners implement Acumatica for food manufacturers in Texas?”
• Compliance: “Which ERP systems support FDA 21 CFR Part 11 for a medical device company?”

Each of these questions points an assistant toward a different kind of source: industry pages, comparisons, the vendor’s own migration documentation, cost guides, partner directories and compliance documentation.

How does an AI answer become an ERP deal?

Through the long list: the answer shapes who is considered, then consultants, demos and partners decide the rest.

Long list. The assistant names a handful of systems for the buyer’s industry and size. Vendors missing from that list may never receive the request for proposals.

Consultant and request for proposals. A selection consultant or internal team narrows the list. Here independent rankings matter, which is why vendors cite analyst reports on their own pages.

Demos and references. Scripted demos and reference calls test the claims. We infer that the AI answer’s influence shows up here not as a tracked referral but as a buyer arriving with a view of your strengths and weaknesses.

Partner selection and contract. The buyer chooses an implementation partner, often from the vendor’s directory. For the vendor, the deal is a multi-year subscription; for the partner, a services engagement. Both depend on being on the original list.

Why does an assistant put some ERP systems on a long list and leave others off?

The companies behind the assistants disclose little; our studies show them hunting for named rankings, reviews and recent pages.

What Google describes. By Google’s account, AI Mode applies a “query fan-out” technique (opens in a new tab): it issues multiple related searches across subtopics and data sources, then combines the results. A single ERP question can therefore pull in analyst coverage, consultant rankings, comparison sites and vendor pages at once.

Observed in our studies. In our hidden-searches study, ChatGPT looked for reviews in 46.2% of answers, and when one of its searches named a source, the answer cited that source 44.0% of the time, against 8.1% when no search named it. That is an association, but it suggests that being the source an assistant searches for by name, such as a well-known ranking, matters. Our study of AI citations and Google rankings found only 8.3% of the pages ChatGPT cited ranked in Google’s top 10 for the question, so strong search rankings alone do not secure a place in the answer. And our freshness study found the assistants cited pages first published about half as long ago as Google’s top 10 for the same questions (a ratio of 0.50), which favors current editions of rankings and recently updated guides.

Trust factors specific to ERP. Buyers weigh industry fit, analyst and consultant assessments, implementation track record, partner coverage in their region, total cost of ownership and references from similar companies. A reasonable expectation is that assistants answering ERP questions lean on the same public evidence: Gartner Magic Quadrants as summarized by vendors and press, consultant reports such as Panorama’s, comparison sites such as Top10ERP, partner directories and detailed case studies.

What does a missed long list cost an ERP vendor?

The whole evaluation, usually for many years.

Because ERP is replaced rarely, a company that chooses a rival in 2026 may not return to the market for a decade. We infer that missing the long list during the 2027 SAP migration window, or during a growth company’s first ERP purchase, costs the vendor not only that contract but also the renewals, expansion modules and partner services that follow. And because a single consultant can carry the same long list into several projects, an omission can repeat. For the broader link between AI answers and pipeline, see our article on AI answers and pipeline.

What does GEO involve for an ERP vendor selling through consultants and partners?

It makes your industry fit, costs and partners easy to verify, without promising a place on any long list.

In ERP, generative engine optimization (GEO) comes down to six workstreams, one of them run with your implementation partners:

1. Industry pages with specifics. Publish pages for each industry and company size you serve, with modules, compliance features and named customer examples, so an assistant can match you to a specific question.
2. Independent assessments, kept current. Participate in analyst evaluations and consultant rankings, and keep summaries of the latest editions on your site. How that kind of third-party standing is built is the subject of how brands build authority for AI search.
3. Migration and comparison guides. Publish honest guides for buyers leaving SAP ECC, QuickBooks or legacy systems, and fair comparisons. For the evidence on that format, read our article on comparison pages.
4. Implementation facts. State typical timelines, cost ranges and what is included, because buyers ask, and Panorama warns that costs left out of estimates are where ERP budgets go wrong.
5. A clean partner network. Keep your partner directory public, current and consistent with partners’ own sites, so answers to “who implements this near me” are right.
6. Tracking by stage of the selection. Re-run industry-fit, comparison, migration, cost and partner questions in ChatGPT, Gemini, Perplexity, Copilot and Google, and tie the results to the requests for proposals you receive and your pipeline; how to design AI visibility tracking explains the setup.

What don’t we know yet about AI’s role in ERP selection?

Nobody has measured how often assistants shape ERP long lists, or what that influence is worth.

The ERP buying data here is either older (Software Path’s 2022 figures) or vendor-reported (SAP, Microsoft, Odoo, Top10ERP). The Gartner survey covers business buying in general, not ERP. Our own studies record what assistants search for and cite across several industries, with no ERP-only sample and no view of which system a buyer then signs for. Treat AI as a real and growing influence on who gets considered, with its effect on ERP contracts still unmeasured.

How can an ERP vendor learn whether it is missing from AI-built long lists?

Ask assistants the questions your target industries ask, and see which systems they name and why.

Run industry-fit, comparison, migration, cost and partner questions through the main assistants and Google’s AI features, then line the answers up against the requests for proposals you received and the deals you won, so you can see which long lists you are missing before a request for proposals goes out. To work through that comparison with us, and plan how to close the gaps with your consultants and partners in mind, talk to us about an ERP visibility audit. Our generative engine optimization service page spells out the ongoing work for an ERP vendor, including industry pages, migration guides and a partner directory that matches partners’ own sites.

Frequently asked questions

Do enterprise ERP buyers really use ChatGPT to choose a system?

Many business buyers now use generative AI to gather vendor information, but they check what it says with people. For ERP, we expect AI to shape the first long list more than the final choice.

Do analyst reports still matter if buyers ask AI?

Yes, and possibly more. Assistants in our study often searched for named rankings and publications, and cited the named source far more often when they did.

Should implementation partners care about AI visibility?

Yes. Buyers ask assistants who implements a given system in their industry and region, and the answer draws on partner directories and partners’ own sites. Consistent, current partner information helps both the vendor and the partner.

How can an ERP vendor tell if AI is influencing its pipeline?

Ask new prospects and consultants how they built their long list, track AI referrals to industry and pricing pages, and regularly check which vendors assistants name for your buyers’ questions. Then compare that with requests for proposals received and lost.

Sources

• SAP (2020), SAP S/4HANA Maintenance Until 2040: Clarity and Choice for SAP Business Suite 7 (opens in a new tab)
• SAP (2026), SAP ERP (opens in a new tab) and RISE with SAP (opens in a new tab)
• Microsoft (2026), FY26 Q4 earnings (opens in a new tab)
• Software Path (2022), ERP Software Report (opens in a new tab)
• Panorama Consulting Group (2026), ERP Software Research and Reports (opens in a new tab) and The Real Cost of Implementing an ERP System (opens in a new tab)
• Top10ERP (2026), Top10ERP (opens in a new tab)
• Odoo (2026), Odoo (opens in a new tab)
• Gartner (2026), Gartner Survey Finds 69% of B2B Buyers Turn to Sales Reps to Validate AI-Generated Insights (opens in a new tab)
• Google (2025), Expanding AI Overviews and introducing AI Mode (opens in a new tab)
• Underneath (2026), hidden searches by AI assistants, AI citations and Google rankings and source freshness