---
title: "How industrial software vendors win plant pilots via AI search"
description: "When plant teams research MES, maintenance or industrial data platforms with AI, the vendors named are those whose plant-level facts are easy to verify."
canonical: "https://underneath.agency/resources/industrial-software-plants-ai-search"
published: 2026-10-07
updated: 2026-10-08
publisher: "Underneath (https://underneath.agency/agent)"
entity: "https://underneath.agency/.well-known/entity.json"
---
Guide · AI search

# Will plant leaders find our MES or IIoT platform when they ask AI for options?

They will if an assistant can find and confirm what your platform does on a real shop floor: which processes, which equipment, which industries and how fast a pilot pays back. Industrial software is still a fragmented, under-adopted market, so many plants are choosing a first system now, and B2B buyers increasingly start that research with generative AI.

This guide is for companies that sell industrial software and industrial IoT (IIoT) platforms to plants: manufacturing execution systems (MES), predictive maintenance and asset performance tools, industrial data platforms and digital twin software. It is not about robots or the integrators who install automation. The prize is a pilot at one plant that can grow into a contract across many.

## The short version

1. Most plants have no commercial system yet: [IoT Analytics](https://iot-analytics.com/mes-vendors-replace-pen-paper-spreadsheets/) estimates 54% of plants worldwide ran operations on pen, paper or spreadsheets in 2024, and just 8% used a commercial MES.
2. The market is crowded and split by vertical: more than 300 vendors serve a $5.5 billion MES market, the leader holds less than 10% share, and nine different vendors lead the 13 industries IoT Analytics analyzed.
3. Budgets are there: in [Rockwell Automation’s 2025 State of Smart Manufacturing Report](https://rockwellautomation.com/en-us/company/news/press-releases/Ninety-Five-Percent-of-Manufacturers-Are-Investing-in-AI-to-Navigate-Uncertainty-and-Accelerate-Smart-Manufacturing.html), 95% of 1,560 manufacturers had invested or planned to invest in AI and machine learning within five years.
4. Large plants commit real money: in [Deloitte’s 2025 Smart Manufacturing Survey](https://www.deloitte.com/us/en/insights/industry/manufacturing/2025-smart-manufacturing-survey.html) of 600 executives, 78% put more than 20% of their improvement budget into smart manufacturing, and execution systems were a top-two investment priority for 33%.
5. Decisions run through pilots: 41% of organizations in one 2025 survey were piloting digital twins, against 20% with full integration, so the vendor that gets on the longlist gets the chance to prove value on one line.

## Who buys industrial software for a plant, and what is a customer worth?

Operations leaders usually own the decision, with IT, OT engineers, integrators and procurement shaping the shortlist.

Deloitte found that 51% of large manufacturers say smart manufacturing initiatives are owned by operations leaders such as the chief operating officer, and 38% by technology owners such as the chief technology officer. Around them sit plant managers, controls and OT engineers, reliability and quality leaders, and procurement. System integrators matter too: IoT Analytics notes that many MES projects include a large share of integrator-led customization. How plants find those integrators is covered in [how automation integrators win industrial buyers](https://underneath.agency/resources/automation-integrators-industrial-buyers-ai-search).

The market these buyers face is large and unsettled. IoT Analytics puts MES at about 6.5% of an $85 billion industrial software market, and counts roughly 5 million factories worldwide. Its [digital twin research](https://iot-analytics.com/our-coverage/iot-platforms-software/) sizes the standalone digital twin market at $1.3 billion in 2025, forecast to reach $4.2 billion by 2030. With only 8% of plants on a commercial MES, the competition for most plants is not a rival vendor. It is a spreadsheet, a homegrown tool or an add-on to the existing ERP.

There is no public benchmark for the value of a typical MES or IIoT contract, so we do not quote one. The public data shows the shape of the opportunity instead:

- Deloitte’s respondents were companies with at least $500 million in revenue, and nearly nine in ten expected their smart manufacturing investment to continue or increase in the next fiscal year.
- Research from Siemens and S&P Global, [reported by Process Excellence Network](https://www.processexcellencenetwork.com/tools-technologies/news/manufacturing-downtime-operational-costs-digital-twins), found almost a third (30%) of organizations spending over $10 million on digital twin technology, in a study of 907 businesses.
- Deloitte’s respondents reported, on average, a 10% to 20% improvement in production output from smart manufacturing. That is the business case your champion has to defend internally.

Our inference: the value of a new customer sits less in the first license and more in the rollout. A platform proven on one line can spread to every line, then every plant.

## Where do AI assistants enter a plant software decision?

Early, when teams frame the problem and build a longlist, and then again when they check what they were told.

We found no survey that isolates how plant teams use AI to choose industrial software. The closest evidence is cross-industry. In [Gartner’s survey of 645 B2B buyers](https://www.gartner.com/en/newsroom/press-releases/2026-05-20-gartner-survey-finds-sixty-nine-percent-of-b-two-b-buyers-turn-to-sales-reps-to-validate-ai-generated-insights), 45% said they used generative AI in a recent purchase, primarily to gather information on vendors and products, and buyers used an average of seven information sources. The same survey found 69% prefer to validate AI-generated insights with sales reps. AI starts the research. It does not finish it.

Manufacturers are also adopting AI inside the plant, which makes them comfortable using it for research. Rockwell found 50% of manufacturers planned to apply AI to product quality in 2025, the top use case for the second year running. Deloitte found 24% had deployed generative AI at facility or network scale, and 23% were piloting AI and machine learning.

The traditional channels still matter. IoT Analytics sent analysts on more than 400 booth visits at [Hannover Messe 2025](https://iot-analytics.com/?p=161547), a reminder that trade shows, analyst reports and integrator advice remain part of how plants find vendors. We infer that AI answers increasingly sit in front of all of these, because an assistant summarizes trade press, analyst coverage and vendor pages in a single reply.

## Which plant questions do buyers put to AI?

Questions that describe a production problem, a vertical and an existing system, rather than a product name.

The plant questions below are our own illustrations of that pattern, not queries logged from real buyers.

| Buyer’s situation | Example question |
|---|---|
| Replacing paper | “MES for a mid-size food plant that still tracks batches on spreadsheets” |
| Downtime | “Predictive maintenance software for pumps and motors that works with our existing vibration sensors” |
| Data foundation | “Industrial data platforms that support a unified namespace and connect older PLCs” |
| Regulated production | “MES vendors with electronic batch records for pharmaceutical plants” |
| Digital twin | “Digital twin software for a packaging line where we can run a pilot in under six months” |
| Replacing a vendor | “Alternatives to our current MES for discrete assembly with better cloud options” |

Two things stand out. Vertical fit is the first filter: IoT Analytics argues that “one size fits all” is unrealistic in MES and that vendors should focus on specific verticals. And integration is the second: the buyer already owns controllers, historians and an ERP, so the question is whether your platform connects to them.

Deloitte’s priority list shows where demand is concentrated. Over the next two years, respondents ranked advanced production scheduling (35%), execution systems (33%) and quality management (28%) as their first or second system investment priorities, with 40% naming data analytics among their top solution priorities.

## How does an AI answer become a pilot and then a multi-site contract?

Through a longlist, a demo, a pilot on one line, a measured result and a rollout across plants.

1. **Framing.** An operations or engineering lead asks what kind of system fixes the problem, and which vendors do it in their industry. The answer shapes the category and the longlist.
2. **Checking.** The team reads vendor sites, analyst coverage, integrator recommendations and peer references. Few teams rely on a single source.
3. **Piloting.** One line or one plant tests the platform. Pilots are normal here: the Manufacturing IT/OT Trend Report 2025 found 41% of organizations in the pilot phase with digital twins and 20% fully integrated.
4. **Proving value.** The pilot is judged on downtime, quality, throughput or labor. Of organizations that had used digital twins, 65% reported reduced downtime and operational costs and 55% improved predictive maintenance.
5. **Rolling out.** A successful pilot becomes a program across lines and sites, with recurring subscription revenue.

AI visibility influences the first two steps. Your product, implementation team and integrators win the last three. A platform that is missing from the longlist never gets the pilot.

## What gets an MES or analytics platform named by an assistant?

Facts it can find and confirm in trusted sources; platforms document how they search, not how they rank.

**Documented by the platforms.** [OpenAI says](https://help.openai.com/en/articles/9237897-chatgpt-search) ChatGPT search rewrites a question into one or more targeted queries sent to search providers, and that a site must allow its crawler, OAI-SearchBot, to be eligible for inclusion. [Google says](https://blog.google/products/search/ai-mode-search/) AI Mode uses a “query fan-out” technique that issues multiple related searches across subtopics, and its [guidance for site owners](https://developers.google.com/search/docs/appearance/ai-features) says there are no additional requirements to appear in AI Overviews or AI Mode beyond being eligible for Search. Neither company publishes how it chooses which vendors to name.

**Observed in our studies.**

- Behind each buyer question sit several searches: [our hidden-searches study](https://underneath.agency/research/ai-hidden-searches-study) counted a mean of 3.7 per question before ChatGPT answered. A question about MES for food plants may be answered from searches about traceability, batch records and cloud MES vendors.
- Google rank is a weak proxy here: [our study of AI citations and Google rankings](https://underneath.agency/research/ai-citations-google-rankings-study) found just 8.3% of the pages ChatGPT cited sat in Google’s top 10 for the question. Ranking for your category term is not the same as being cited.
- In [our crawler study](https://underneath.agency/research/ai-crawler-blocking-study), 7.4% of top sites blocked OpenAI’s search crawler. A security-minded IT team can block it without realizing what that does to visibility.

**Our inference for industrial software.** The trust factors are the ones a plant team already checks, written where machines can read them: industries and processes served, equipment and protocols supported, deployment options (cloud, edge or on-premises), OT security posture, typical pilot scope and time to value, named reference plants used with permission, integrator partnerships, analyst coverage and trade press. If those facts live only in gated PDFs and sales decks, an assistant has little to quote.

## What does an industrial software vendor lose when it is left out?

Pilots it never hears about, in a market where most plants are choosing their first system.

We have no measurement of pilots lost to AI absence, so we label the reasoning as ours:

- **The buying window is open now.** With 54% of plants on paper or spreadsheets and Rockwell reporting that 81% of manufacturers say outside and internal pressures are speeding up digital transformation, many plants are making a first selection rather than renewing an incumbent.
- **The field is wide.** Three hundred MES vendors cannot all make a longlist. When an assistant names four or five, the rest are not compared.
- **Losses are invisible.** A plant that piloted a rival never appears in your pipeline, so no report records the miss.
- **Wrong facts filter you out.** An answer that says you lack an on-premises option, or do not serve regulated industries, removes you from a pharma or defense evaluation. To find which page planted the error and get it changed, use our guide on [correcting wrong brand facts in AI answers](https://underneath.agency/resources/fix-wrong-brand-information-in-ai-answers).

## How does GEO work for an industrial software or IIoT company?

Generative engine optimization (GEO) makes your platform easy for AI assistants to find, describe accurately and verify.

For an industrial software vendor, the work usually covers:

1. **Vertical pages written as plain text.** One page per industry and use case, stating processes, regulations, equipment and outcomes, not just a brochure download.
2. **Integration facts.** Supported controllers, historians, ERP connectors and protocols such as OPC UA and MQTT, listed in text a crawler can read.
3. **Deployment and security facts.** Cloud, edge and on-premises options, data residency and OT security practices, because IT and security teams join the evaluation.
4. **Proof from plants.** Reference sites, measured results and case studies published with customer permission, with the plant’s industry and scope stated.
5. **Independent coverage.** Analyst reports, trade publications, conference talks and integrator partner directories. Our article on [how brands build authority for AI search](https://underneath.agency/resources/how-brands-build-authority-for-ai-search) explains why outside sources carry weight.
6. **Consistent identity.** The same product names, capabilities and partner lists on your site, marketplaces, partner pages and review platforms, especially after acquisitions and rebrands.
7. **Crawl access.** Allow the search crawlers the assistants document, and keep key pages out from behind forms.
8. **Measurement.** Ask a fixed set of plant-problem questions across ChatGPT, Gemini, Perplexity, Copilot and Google’s AI features, repeatedly, and record who is named and which sources are cited. Then compare that with where pilots actually come from.

No vendor can be promised a place on a plant’s longlist, and anyone who offers that guarantee is overselling. What this work does is make your platform the easiest one for an assistant, and then an engineer, to check. Software buyers in other categories follow a similar path; see [how ERP vendors get on AI-built long lists](https://underneath.agency/resources/erp-software-ai-search) and [whether AI assistants shape enterprise software shortlists](https://underneath.agency/resources/enterprise-software-shortlists-ai-search). Engineering teams choosing design tools face a similar test, covered in [how CAD and simulation vendors get recommended](https://underneath.agency/resources/engineering-software-ai-search).

## Which questions can’t the data answer yet for industrial software?

It shows manufacturers investing and buyers using AI, not how many pilots AI answers create.

- **No plant-specific AI buying data.** The AI usage figures here come from cross-industry B2B surveys. We found no public survey of how plant teams use AI to choose MES or IIoT software.
- **Sponsors have interests.** Rockwell sells automation and software, Siemens sells digital twin software, and IoT Analytics sells market reports to vendors.
- **No ranking studies for this category.** Our studies covered buyer questions across several industries. Applying them to industrial software is our inference.
- **Contract values are private.** Public figures show budgets and market sizes, not the typical value of a plant subscription or a multi-site program.

## Where should an industrial software company start?

Ask assistants the plant problems your best customers had before their first pilot, and see whether you are named.

That first check usually shows whether your platform appears for its core use cases and verticals, whether integrations and deployment options are described correctly, which analyst, trade and integrator sources the answers rely on, and which rivals are named instead.

If your growth depends on turning a handful of pilots into multi-site programs each year, [talk to us about a visibility review](https://underneath.agency/contact). We will show where your platform appears when plant teams ask AI for options, why rivals are named instead, and which fixes are most likely to put you on more pilot shortlists. Our [generative engine optimization service](https://underneath.agency/services/generative-engine-optimization) page sets out how vertical pages, integration facts and plant proof are built and measured for MES and IIoT vendors.

## Frequently asked questions

### Do plant managers really use ChatGPT to research MES or IIoT vendors?

There is no plant-specific survey yet. Cross-industry surveys show many business buyers use generative AI to gather vendor information, then check it with people and other sources.

### Should we publish integration and pricing details publicly?

Publish integration, deployment and industry facts in plain text. Pricing is a business decision; at minimum, say how pilots are scoped and what drives cost, so an assistant does not guess.

### Do analyst reports on MES still count when plant teams ask AI?

Yes. Assistants cite outside sources, and analyst coverage, trade press and integrator directories are among the sources plant teams already trust. We infer they also shape what assistants say.

### Can a niche vendor compete with the largest automation companies?

In vertical questions, often yes. The MES leader holds less than 10% share, and nine different vendors lead the 13 industries IoT Analytics studied. Specific facts help a specialist win the specific question.

## Sources

- IoT Analytics (2025-12-15), [Manufacturing Execution Systems: The 300+ vendors looking to displace pen, paper, and spreadsheets in the factory](https://iot-analytics.com/mes-vendors-replace-pen-paper-spreadsheets/)
- IoT Analytics (2026-08-26), [Inside the $1.3B digital twin market (IoT platforms and software coverage)](https://iot-analytics.com/our-coverage/iot-platforms-software/)
- IoT Analytics (2025-05), [Hannover Messe 2025: the latest Industrial IoT/Industry 4.0 Trends](https://iot-analytics.com/?p=161547)
- Rockwell Automation (2025-06-03), [Ninety-Five Percent of Manufacturers Are Investing in AI to Navigate Uncertainty and Accelerate Smart Manufacturing](https://rockwellautomation.com/en-us/company/news/press-releases/Ninety-Five-Percent-of-Manufacturers-Are-Investing-in-AI-to-Navigate-Uncertainty-and-Accelerate-Smart-Manufacturing.html)
- Deloitte (2025), [2025 Smart Manufacturing and Operations Survey](https://www.deloitte.com/us/en/insights/industry/manufacturing/2025-smart-manufacturing-survey.html)
- Process Excellence Network (2025-04-23), [Manufacturing firms reduce downtime and operational costs with digital twins](https://www.processexcellencenetwork.com/tools-technologies/news/manufacturing-downtime-operational-costs-digital-twins)
- Gartner (2026-05-20), [Gartner Survey Finds 69% of B2B Buyers Turn to Sales Reps to Validate AI-Generated Insights](https://www.gartner.com/en/newsroom/press-releases/2026-05-20-gartner-survey-finds-sixty-nine-percent-of-b-two-b-buyers-turn-to-sales-reps-to-validate-ai-generated-insights)
- OpenAI Help Center (2026), [ChatGPT search](https://help.openai.com/en/articles/9237897-chatgpt-search)
- Google (2025-03-05), [Expanding AI Overviews and introducing AI Mode](https://blog.google/products/search/ai-mode-search/)
- Google Search Central (2025), [AI features and your website](https://developers.google.com/search/docs/appearance/ai-features)
- Underneath (2026), [The hidden searches AI assistants run before they answer](https://underneath.agency/research/ai-hidden-searches-study)
- Underneath (2026), [Do ChatGPT, Gemini, Perplexity and Claude cite pages that rank?](https://underneath.agency/research/ai-citations-google-rankings-study)
- Underneath (2026), [Which AI crawlers do top websites block?](https://underneath.agency/research/ai-crawler-blocking-study)

---

This is the Markdown twin of https://underneath.agency/resources/industrial-software-plants-ai-search. The HTML page is canonical. Publisher: Underneath, https://underneath.agency/agent. Site index: https://underneath.agency/llms.txt.