Requirements, evaluation criteria, and red flags that separate AEO operators from content mills in 2026.

Most AEO RFPs in 2026 are recycled SEO briefs with AI Overviews pasted into the KPI line. Content mills answer them in a weekend. A real AEO RFP names citation KPIs by engine, ownership of entities and prompts, and a pilot that proves extractable commercial pages before a retainer.
Here is the shape: requirements that kill mill bids, evaluation criteria, a scoring rubric, red flags, and pilot deliverables. Lock the delivery model first with AEO agency vs AEO consultant. Scope lives on AEO for B2B and AEO for B2B SaaS.
G2's 2026 AI Search Insight Report, fielded among 1,076 B2B software buyers in March 2026, found 51% now start research with an AI chatbot more often than Google, and 69% chose a different vendor than expected because of chatbot guidance. Hire for that surface.
Require named people on Slack, a prompt log across AI Overviews, AI Mode, ChatGPT, and Perplexity, and citation share against your competitor set. Require extractable commercial pages. Google's AI features guidance says pages need to be indexed and snippet-eligible; there are no extra technical requirements. Require ownership of prompts, entities, and dashboards. Measure Overviews in Search Console and assistants with GA4 plus a prompt log (track AI Overviews, measure AEO results).
As of March 2026, Ahrefs found only ~38% of AI Overview citations also ranked top 10 for the same query across 863K SERPs and 4M URLs (Ahrefs, March 2, 2026). Rankings alone are not AEO proof.
Book a 30-minute call if you want a second set of eyes on the RFP before vendors see it.
Use a 1-5 scale. Two reviewers. Reject below 3 on engine fluency or named operators. Cap content production capacity at 10% of score.
| Score | Engine fluency | Measurement |
|---|---|---|
| 5 | Different tactics for Overviews, AI Mode, ChatGPT, Perplexity | Prompt log + GSC + GA4 + competitor set |
| 1 | Treats AEO as classic SEO | Cannot describe a prompt log |
Ask for 60-90 days on 20-40 commercial queries, a baseline citation audit, three to five money-page rewrites, entity cleanup, and weekly citation-share by engine. The Free AEO Report Card is Austin Heaton's diagnostic of citation gaps and money-page readiness; use a baseline like it before the pilot. Pay for the pilot. Retainers: pricing.
Book a 30-minute call to turn this into a vendor scorecard.
Austin Heaton's take: the RFP is a filter, not a wish list. If a vendor cannot name engines, own a prompt log, and ship extractable commercial pages in a paid pilot, they are a content mill with new vocabulary. The downside is fewer bids and more time writing requirements. For teams where G2's 51% already start in a chatbot, that trade is correct. Start with AEO for B2B SaaS or AEO for B2B if you need the operator model.
Write the requirements this week. Score with two reviewers. Run a paid pilot before the retainer. Book a call to pressure-test the RFP.
Named operators, a multi-engine prompt log, citation KPIs, ownership of entities and dashboards, extractable commercial pages, and a paid 60-90 day pilot.
Cap content volume weighting, require engine-specific tactics, name people in the SOW, reject ranking guarantees, and require citation share baselines.
Add citation share by engine and money-page extractability. As of March 2026, Ahrefs finds only ~38% of AI Overview citations also rank top 10 for the same query.
Yes for first-time AEO hires. Pay for a 60-90 day pilot on a fixed prompt set.
Lock the delivery model before scoring. See AEO agency vs AEO consultant.
The Free AEO Report Card is a diagnostic of citation gaps and money-page readiness. Attach a baseline like it so every vendor scores the same prompt set.