ERP in ChatGPT: Twelve Vendors Named, Zero Sources, One Comparison Site Deciding the Shortlist

ERP and manufacturing is one of the weakest categories in this study for vendor control. Vendors took 52.5% of citations, against 90.9% in accounting. In one capture, ChatGPT named twelve ERP vendors – including Oracle, SAP, Microsoft and Epicor – and not one of them had a supporting citation to its own website. Five of the seven citations in that answer came from a single small comparison site with no ownership disclosure. If you sell enterprise ERP, a large share of your AI citation surface currently belongs to publishers you have never heard of.

The category at a glance

Seventeen distinct brands surfaced across six prompts, with five different vendors taking the top slot. SAP S/4HANA was the most frequent leader and held it only 33% of the time. There is no stable "ChatGPT recommends X for ERP" answer.

The capture that should worry every ERP vendor

Prompt (P46, alternatives-to): "What are the best alternatives to NetSuite?"

ChatGPT read 34 results across nine domains and rendered seven citations. Five came from erpresearch.com. Two came from cudio.com. Answer latency: 10.4 seconds.

Twelve brands were named in the answer: Oracle, NetSuite, Microsoft Dynamics 365 Business Central, Acumatica, SAP Business One, Odoo, Microsoft Dynamics 365, Sage X3, Epicor, IFS, Oracle Fusion and SAP S/4HANA.

Not one of those twelve had a citation to its own site anywhere in the answer.

So the shortlist a buyer sees for a serious, multi-year, seven-figure enterprise decision rests on a single third-party comparison page plus whatever the model already believes. Oracle, SAP and Microsoft between them contributed nothing to the evidence base of an answer that recommended all three.

erpresearch.com publishes ERP rankings and head-to-head comparisons. It states that "no vendor pays for placement" and discloses no ownership, no staff and no revenue model. Across the full sixty-prompt corpus it earned 11 citations in four answers, making it the fifth most-cited domain in the entire study – ahead of Salesforce, Microsoft, Asana and Gartner.

Why ERP behaves this way

The mechanism is pricing. ERP pricing is gated behind sales conversations, which removes the single most citable artefact a vendor owns.

Across the study, categories with public pricing gave vendors 80–91% of citations. Categories with gated pricing gave vendors 48–55%. ERP sits at 52.5%, squarely in the gated band, alongside healthcare EHR at 48.7% and legal practice management at 54.1%.

Citations bind to checkable claims. If the vendor will not supply a price, tier structure or feature matrix in extractable form, the model takes those facts from whoever will – and in ERP, that is a small set of specialist comparison sites and analyst pages.

The consequence shows in citation concentration too. ERP's citation HHI is 1,312 across 13 cited domains, against 2,213 across nine domains in accounting. Where vendors do not supply the evidence, the citation surface spreads across third parties, and no single one of them can be managed.

Which ERP brands the model already believes in

Microsoft Dynamics 365 has the second-broadest presence of any brand in the entire sixty-prompt corpus – nine answers, behind only HubSpot's eleven – and led none of them. Mean rank 4.44. It is the archetypal "always on the list, never the pick" position.

NetSuite is similar at mean rank 5.20 across five answers with zero top slots. Epicor sits at 5.80, the weakest mean rank of any brand in the most-recommended table.

Compare those to the leaders in transparent categories – Clio at 1.20, Shopify at 1.50, Asana at 1.60 – and the pattern is clear. In gated categories the model names many vendors, ranks them weakly, and holds no strong conviction about any of them.

What this means if you sell ERP

You are competing for citation share you have already forfeited. 47.5% of citations in this category go to non-vendor sources. Some of that is legitimate analyst work. A meaningful share is small comparison sites publishing dated tables.

Publish structured, checkable detail even if you cannot publish a price. The choice is not between a fixed number and silence. Publish the pricing model – per-user, per-module, per-transaction, tiered by volume. Publish a band or a starting point. Publish seat minimums, module inclusions, implementation scope, supported integrations, supported deployment models, industry-specific configurations. Every one of those is a claim a citation can bind to, and most ERP vendors bury all of them in a gated PDF.

Audit the comparison sites currently defining your category. If a single page is supplying five of seven citations in an answer about NetSuite alternatives, what that page says about your product is a commercial issue rather than a PR one. Most of these sites have a contact route and an incentive to be accurate. This is one of the highest-leverage actions available in a gated category.

Build the alternatives-to page yourself. Alternatives-to prompts named a mean of 7.8 brands per answer across the corpus, second only to category-landscape questions at 10.9. For a challenger ERP vendor, that archetype is the widest available entry into a shortlist – and if you do not publish "alternatives to NetSuite", erpresearch.com will.

Accept that mean rank moves slowly. Microsoft's 4.44 across nine answers is an entity-salience position, not a content problem. 69.3% of brand recommendations corpus-wide carried no citation at all, so a large share of ERP ranking is decided before retrieval begins. Work the citation half quarterly and treat the ranking half as a multi-year programme.

FAQ

Does ChatGPT cite ERP vendors' own websites?

Less than in most categories. ERP and manufacturing vendors took 52.5% of citations, against a 65.9% corpus average and 90.9% in accounting. The gap goes to analysts, trade press and specialist comparison sites, driven mainly by gated pricing.

What ERP software does ChatGPT recommend?

It varies heavily by phrasing. Across six ERP prompts, five different vendors took the top slot and SAP S/4HANA - the most frequent leader - held it only 33% of the time. Seventeen distinct brands surfaced in total, with Microsoft Dynamics 365 appearing in nine answers across the wider corpus without ever ranking first.

Why did ChatGPT name twelve ERP vendors without citing any of them?

Because the answer's shortlist came from the model's prior belief, not from retrieval. In that capture, seven citations were rendered from just two third-party domains, and citations attached to the comparison source rather than to the twelve recommended products. Corpus-wide, 69.3% of brand recommendations carry no citation to that vendor.

Which sites does ChatGPT cite for ERP comparisons?

In this corpus, erpresearch.com was the dominant source, with 11 citations across four answers - the fifth most-cited domain in the entire study. It publishes ERP rankings and head-to-head comparisons and discloses no ownership, staff or revenue model.

How can enterprise ERP vendors improve AI citation share without publishing prices?

By publishing everything adjacent that is not negotiable: the pricing model, a starting band, seat minimums, module inclusions, implementation scope, supported integrations, deployment models and industry configurations. Each is a checkable claim a citation can bind to.

Related in this series

• How ChatGPT Shortlists Software Brands, the full report as a PDF
• Forty-Five Reads, Six Links: How ChatGPT Actually Cites Software Brands
• The Sites Beating G2 in ChatGPT: 43 Comparison Sites, 53 Citations, 3.5x the Aggregators
• 69.3% of ChatGPT's Software Recommendations Have No Source Behind Them
• Publish Your Prices: Why Gated Pricing Costs You 30 Points of AI Citation Share

About the research. Nathan Mzumara is an organic growth and AI search practitioner. Category figures rest on six observations and should be read as directional. Method and limitations are stated in the pillar report.