AEO WRANGLER B2B SOFTWARE INDEX

Shop Floor Execution Visibility Index

Which MES and MOM vendors AI answer engines actually name when discrete manufacturers ask what to buy

112 prompts · 8 buyer personas · ChatGPT, Gemini, Perplexity · 336 responses · Run September 2026

Download the full report (PDF, 21 pages)

Core MES and MOM vendors · 112 prompts · Run September 2026
RankVendorScoreChatGPTGeminiPerplexityUsage wtd.Personas
1Siemens Opcenter59.264.364.349.161.98 / 8
2Rockwell (Plex + FactoryTalk)40.547.337.536.642.98 / 8
3Tulip37.531.244.636.635.98 / 8
4Critical Manufacturing36.352.738.417.943.18 / 8
5Dassault DELMIA26.534.821.423.229.28 / 8
6SAP Digital Manufacturing24.427.722.323.225.58 / 8
7Aegis FactoryLogix15.520.516.19.817.67 / 8
8GE Vernova Proficy11.917.99.88.014.07 / 8
9AVEVA MES11.315.26.212.512.26 / 8
10MasterControl Mx10.49.813.48.010.55 / 8
11Applied Materials SmartFactory10.17.112.510.79.25 / 8
12iBase-t Solumina8.09.88.95.48.93 / 8
13First Resonance ION6.83.614.32.76.43 / 8
14Eyelit6.88.93.68.07.33 / 8
1542Q6.516.10.92.79.75 / 8
16Infor MES / CloudSuite6.23.68.96.25.56 / 8
17iFactory4.50.00.912.52.27 / 8
18iTAC MOM Suite4.58.02.72.75.72 / 8
19Körber PAS-X3.96.22.72.74.73 / 8
20IBM SiView3.96.22.72.74.71 / 8
21Sepasoft MES3.38.90.90.05.35 / 8
22Oracle Manufacturing / Fusion3.33.62.73.63.36 / 8
23Epicor Advanced MES3.33.62.73.63.32 / 8
24Epsilon33.00.04.54.52.03 / 8
25PICO MES2.74.52.70.93.44 / 8
26Factory AI2.70.00.08.01.35 / 8
27WorkCell / ShopVue2.70.94.52.72.25 / 8
28Connektica MES2.70.00.97.11.43 / 8
29Parsec TrakSYS2.40.02.74.51.54 / 8
Second place does not resolve cleanly. Rockwell wins on equal weight through consistency across all three engines. Critical Manufacturing wins on usage weight, ranking second on ChatGPT, but falls to seventh on Perplexity. Tulip sits between them with the most balanced profile.
All figures are the percentage of prompts naming each vendor. This measures AI visibility, not product quality or market share.
KEY FINDINGS

What the data shows

Five findings from 336 responses, 134 products named, and 4,777 source citations.

Siemens is the clear leader

Opcenter is named in 83 of 112 prompts and appears in all eight buyer segments on all three engines. No other vendor clears 50 percent coverage on more than one engine. The lead holds under every weighting method tested, so no methodological choice threatens it. What makes the position unusual is that it rests on training recall and retrievable content simultaneously: Siemens has the largest own-domain citation footprint in the study, with 107 citations on ChatGPT alone.

The three engines are not measuring the same market

Critical Manufacturing is named in 52.7 percent of ChatGPT prompts and 17.9 percent of Perplexity prompts. iFactory, a product most people in this category have never heard of, appears in 12.5 percent of Perplexity prompts and none at all on ChatGPT. Of the 847 domains cited across the study, only 21 were cited by all three engines and 536 were cited by Perplexity alone. A vendor measuring its AI visibility on one engine is measuring roughly a third of the picture, and possibly the wrong third.

Visibility comes from two separate mechanisms

For every vendor mention we checked whether the same engine cited that vendor's own website in the same answer. ChatGPT did so 68 percent of the time, Perplexity 24 percent, Gemini 9 percent. That test separates vendors the models remember from vendors the models can find. Some names appear in dozens of prompts with almost no supporting citation anywhere, which means their visibility rests entirely on training data. That is a durable position and also one that publishing cannot improve.

Buyer segment changes the answer more than anything else

Five vendors that sit outside the overall top nine lead or place second in a single segment. MasterControl ranks tenth overall and first in medical device. Aegis ranks seventh and first in electronics. Applied Materials ranks eleventh and second in semiconductor. iBase-t ranks twelfth and second in aerospace. First Resonance ranks thirteenth and second in hard tech. Each of them appears in 11 to 13 of the 14 prompts in their home segment and almost nowhere else.

Adjacent categories take a quarter of the answers

Only 62 percent of product mentions were core MES or MOM products. ERP took 17.6 percent and PLM 6.7 percent. SAP S/4HANA ranks seventh across all products and Siemens Teamcenter eighth, ahead of every specialist MES vendor except the top six. The competitive set in AI answers is wider than the competitive set in a normal deal cycle, because the engines reproduce the buyer's existing mental model, and that model already has ERP and PLM in it.

FAQ

Where does my company rank?

The tables above carry every ranked vendor. Core MES and MOM shows all 29, including adjacent shows all 57 products named in seven or more prompts, and the segment view shows the top ten in each of the eight buyer personas. A vendor's position often differs between those views, which is usually the most useful thing to know: five vendors outside the overall top nine lead or place second in a single segment. If your company is not in any table, it was named in fewer than seven of the 112 prompts, and the full report lists everything below that floor.

Why is my company missing entirely?

Products named in fewer than seven of the 112 prompts are not ranked, because at that volume the difference between two appearances and one is noise rather than a competitive gap. Twenty-four further core products cleared at least one mention but fell below that floor, and they are listed in the full report. If a product was never named at all, that is the finding rather than an omission: 134 products were named across 336 responses, so the engines were not short of names to offer.

How can a vendor I have never heard of rank ahead of one I have?

Because this measures AI visibility, not market position. A vendor can be widely deployed and rarely named, and a small vendor with strong published content can be named often. In this index one vendor competing actively for real deals appeared in 2 of 112 prompts, while another that few people in the category would recognize appeared in 15. Both results are accurate descriptions of what the engines said.

Why are ERP and PLM products in a shop floor ranking?

Because the engines put them there. The extended table reports everything named in response to shop floor execution questions, and only 62 percent of mentions were core MES or MOM products. ERP took 17.6 percent and PLM 6.7 percent. The core table applies a system-of-record test that excludes products serving as the financial system of record, which is why SAP S/4HANA appears in one table and not the other.

Why do the segment scores look so much higher than the overall scores?

Segment scores are out of 14 prompts rather than 112, so they run higher by construction. A vendor scoring 76.2 in hard tech and 37.5 overall is not contradicting itself. Compare segment scores against each other, and overall scores against each other, but not across the two.

Why treat Rockwell Plex and FactoryTalk as one entry?

Where one vendor sells more than one product inside the category, the products count as a single entry, because a buyer choosing between vendors is not choosing between two Rockwell products. This affects Rockwell and Dassault. Combined figures are the union of prompts naming either product rather than the sum of the two. Per-product figures are in the full report.

Working through this for your own company?

Get in Touch →