From invisible to AI-recommended for a smart vending provider
A workplace vending provider that AI models could not see at all — rebuilt into a citable source, producing 446 sessions, 37 key events and a near-tripling of engagement time.
Anonymised at the client’s request — results are real and unedited.
The challenge
The provider supplies smart vending machines, micro-markets, smart fridges and coffee systems to workplaces, handling installation, restocking and maintenance. Despite a genuine product and service offering, it was effectively invisible to AI recommendation engines: no optimised business listings on key platforms, minimal credible mentions or citations across the web, and no consistent brand narrative for a model to draw on. When potential customers asked an AI assistant for a workplace vending supplier, they were pointed at competitors.
Our approach
Content rebuilt around authority signals
Restructured key website content to embed the authority and trust indicators AI models look for when deciding which brands to surface — what the business does, who it serves, and what it is accountable for.
A dedicated llms.txt file
Created and submitted an llms.txt file giving AI models a direct roadmap to the most important pages, so the brand is represented consistently rather than differently on every platform.
Advanced schema markup
Deployed structured data covering the business, its services and its credibility markers in a format AI systems recognise instantly.
FAQs rebuilt from real search behaviour
Researched how people actually search for workplace vending and rebuilt the FAQ section to mirror it, turning it into a high-value answer source models draw from directly.
The results
- 446 sessions from AI platforms, growing quickly and then holding steady — consistent traffic rather than a spike.
- Average engagement time from AI-referred visitors rose from 00:00:28 to 00:01:17 — roughly 2.7x — showing these visitors read rather than bounce.
- 37 key events generated from AI platforms, against a baseline of zero before the work began. That is measurable action, not just traffic.
- Brand presence established across Google AI Overviews, ChatGPT and Perplexity.
Key takeaway
Being absent from AI answers is not a ranking problem you can fix with more content — it is a recognition problem. Listings, citations, structured data and a consistent narrative are what make a business legible to a model in the first place.
Want results like these?
Book a free 15-minute strategy call and we’ll map the highest-leverage moves for your business.
Book a 15-Min Strategy Call →