# Dynamix AI visibility audit

> Dynamix (dynamixinc.com) was named in 23% of 48 answers from ChatGPT, Claude, Gemini and Perplexity to buyer questions that did not mention it (95% range 13–37%, September 2026). Gemini never named it. The site AI linked most instead: mxdprocess.com (10 citations).

- Audited site: https://dynamixinc.com
- Category: Industrial Mixers
- Sampled: 2026-09-14
- Audit by: Ivan Karaman (https://ivankaraman.com)
- Full report: https://ivankaraman.com/audit/industrial/dynamixinc.com

## The numbers

- Unprompted mention rate: 23% of 48 answers (95% Wilson range 13–37%).
- Own site linked as a source: 11 of 48 answers.
- Sample quality: solid.

| Engine | Model | Named | Rate | 95% range |
|---|---|---|---|---|
| ChatGPT | – | 3 of 12 | 25% | 9–53% |
| Claude | – | 4 of 12 | 33% | 14–61% |
| Gemini | – | 0 of 12 | 0% | 0–24% |
| Perplexity | – | 4 of 12 | 33% | 14–61% |

## The buyer questions asked

Each question was asked 4 times on every engine, with web search on. The business name was never in the question.

- "What are the best industrial mixer manufacturers for custom agitators?"
- "Can you recommend a reliable company for process mixing solutions?"
- "Who prices competitively for high-quality industrial mixers?"

## Who AI cited instead

| Site | Type | Times linked |
|---|---|---|
| centennialequipment.com | Other source | 12 |
| mxdprocess.com | Competitor | 10 |
| wmprocess.com | Other source | 9 |
| industrialmixers.com | Other source | 9 |
| iqsdirectory.com | Competitor | 8 |
| liquidhandlingequip.com | Other source | 8 |
| newafluids.com | Other source | 8 |
| mixtec.com | Other source | 7 |
| asctornado.com | Other source | 7 |
| dciinc.com | Media | 7 |
| ekato.com | Other source | 6 |
| miltonroy.com | Other source | 6 |

- **MXD Process**: Deep process-equipment education and mixer product context that aligns with how AI answers application and selection questions.
- **IQS Directory**: Directory taxonomy and supplier blurbs optimized for who-makes-X queries; high reuse in generative shortlists.

## Diagnosis

Generative engines aren't guessing who makes the best industrial mixers; they are synthesizing answers from sources structured for easy retrieval. Right now, Dynamix has an unprompted AI visibility rate of just 22.9% (with a 95% confidence interval of 13.3% to 36.5%). This means you are missing from the majority of AI-generated shortlists when engineers ask brand-blind buyer questions.

The problem isn't a failure of your product—it's how your website frames its information. Gap domains like mxdprocess.com and industrialmixers.com are winning retrieval because they act as process-side equipment education hubs. They publish the exact comparison matrices and viscosity guides that AI models quote when answering application prompts. Distributors like centennialequipment.com provide concrete product language, and directories like iqsdirectory.com offer dense, structured vendor taxonomy optimized for who-makes-X queries.

In contrast, your site's JSON-LD is mostly limited to generic Service, Organization, and WebSite types. It severely under-describes your sellable equipment and lacks the specific Product and FAQPage depth that engines require. This is especially glaring in Gemini, where you received 0 citations out of 12 samples. To correct this, you need to ship a single canonical Industrial Mixer Selection Guide covering impeller types, viscosity ranges, and tank geometry, backed by expanded llms.txt paths and robust Product schema, all without sacrificing your current 89 PageSpeed score.

## Technical readiness

- AI crawlers can read the site: OK. No AI crawler is blocked in robots.txt.
- llms.txt: OK. Present.
- Structured data (JSON-LD): OK. Found Service, Service, Service, Service, Organization, WebSite, ImageObject, WebPage, WebSite.
- Homepage description: OK. Set.
- agents.md: Note. Optional. Very few sites have one yet.

## Method

48 answers: 3 buyer questions × 4 samples × 4 engines, sampled through the engines' APIs with web search on (ChatGPT: unknown, Claude: unknown, Gemini: unknown, Perplexity: unknown). A "mention" is the business named in the answer text; a "citation" is its own URL linked as a source. The two are counted separately. Rates carry 95% Wilson score confidence intervals. Reviewed by hand before publishing.

Source: https://ivankaraman.com/audit/industrial/dynamixinc.com