AEO Agency RFP Template (2025): 24 Questions to Separate Real Capability From Rebranded SEO

Most AEO agency RFPs we've seen are recycled SEO RFPs with "AI" replacing "Google." That approach fails because Answer Engine Optimization requires fundamentally different evaluation criteria: schema depth, citation provenance, multi-platform coverage (Perplexity, ChatGPT, Gemini, Copilot), and content built for machine consumption, not just human readers.

This is our full 24-question RFP template organized into five sections. Each question includes a 1-5 scoring rubric and examples of strong versus weak responses.

Download the .docx template (placeholder)

This template belongs to our RFP cluster. If you're also evaluating agencies for website builds, see our Website RFP Template 2026 Buyer's Guide. For redesigns where SEO matters, we built a Website Redesign RFP Template. Custom software projects need the Software Development RFP Template, which covers technical evaluation beyond what marketing RFPs address.

What Is an AEO Agency RFP?

An AEO agency RFP is what you send prospective Answer Engine Optimization vendors to evaluate how they'd increase your brand's visibility in AI-generated answers across Google's AI Overviews, ChatGPT search, Perplexity, Microsoft Copilot, and similar platforms. Unlike traditional SEO RFPs focused on rankings and organic traffic, an AEO RFP evaluates an agency's ability to get your content cited, quoted, and attributed by large language models.

The structure mirrors the RFP/RFQ/RFI family we outlined in our 7-Step RFP Process guide, but the questions are domain-specific. You're not asking about link-building or keyword research. You're asking about structured data fluency, citation tracking infrastructure, and how they handle AI answers that hallucinate your brand into contexts you'd rather avoid.

Why Generic Marketing RFPs Fail for AEO

Superside's research found that more than one-third of brands receive incomplete or inaccurate responses to creative service RFPs, and nearly 70% of issuers include irrelevant information. Those numbers worsen in AEO, a discipline most agencies have practiced for under 18 months.

The core problem: standard marketing RFPs (like those from HubSpot or Altitude Marketing) ask about campaign scope, timelines, and budget. Necessary, but insufficient for AEO evaluation. They miss the technical layer entirely. When Vital Design's paid advertising RFP template warns about "failure to spend enough time defining campaign scope," they're right—but in AEO, scope definition itself requires technical vocabulary most marketing teams don't have yet.

That's why we built this template. Each question targets a specific capability gap we've seen agencies hide behind jargon.

How to Use This Template

Download the .docx file—placeholder link
Customize the company background section (first two pages) with your brand, industry vertical, and current AEO baseline
Keep all 24 questions intact—they're sequenced to build on each other
Use the 1-5 scoring rubric for each question when evaluating responses
Weight the sections based on your priorities (we suggest Technical Depth at 30%, Tracking at 20%, Content at 25%, Pricing at 15%, Team at 10%)
Send to 3-5 agencies maximum—AEO is niche enough that more than five creates evaluation fatigue without improving outcomes
Give agencies 14-21 calendar days to respond. Less than 14 days filters out agencies actually doing the work (they're busy); more than 21 lets them overthink it.

As we noted in our RFP process guide, the goal isn't collecting the most proposals. It's collecting proposals comparable enough to score.

Section 1: Technical Depth

This section evaluates whether the agency actually understands technical mechanics of how AI engines source, attribute, and surface content. Six questions, weighted at 30% of total score.

Question 1: What schema markup types do you implement for AEO, and how do you decide which types apply to a given page?

Score	What It Looks Like
5	Names specific schema types (FAQPage, HowTo, Article, Organization, speakable) and describes a decision tree based on content type, page intent, and target AI platform. References Schema.org version numbers (e.g., v26.0).
4	Lists schema types with clear rationale but lacks platform-specific nuance.
3	Mentions schema generically, says "we audit and implement structured data." No specifics.
2	Conflates schema markup with meta tags or Open Graph.
1	Doesn't mention schema at all, or says "our dev team handles that."

What good looks like: "For B2B SaaS clients, we typically implement Organization, FAQPage, and Article schema on knowledge base content, with speakable properties on key definition pages. We A/B test schema configurations quarterly against citation rates in Perplexity and Google AI Overviews."

What bad looks like: "We implement all relevant schema markup to ensure maximum visibility."

Question 2: How do you track and verify citation provenance—meaning, when an AI engine cites your client's content, how do you confirm the source and monitor attribution accuracy?

Score	What It Looks Like
5	Describes specific tooling (Profound, Otterly.AI, custom scraping pipelines), explains methodology for matching AI citations back to source URLs, and addresses hallucination monitoring.
4	Uses at least one dedicated AEO tracking tool and has a manual verification process.
3	Says they "monitor AI mentions" but can't explain how.
2	Relies solely on traditional SEO tools (Ahrefs, SEMrush) with no AI-specific tracking.
1	No answer or says citation tracking "isn't possible yet."

Question 3: Which AI answer platforms do you actively optimize for, and how does your approach differ across each?

Score	What It Looks Like
5	Covers Google AI Overviews, ChatGPT (with Browse/Search), Perplexity, Microsoft Copilot, and at least one emerging platform. Explains specific tactical differences (e.g., Perplexity favors direct-answer formatting while Copilot leans on Bing index freshness).
4	Covers 3-4 platforms with some differentiation.
3	Says "we optimize for AI" without naming platforms.
2	Only mentions Google AI Overviews.
1	Treats AEO as identical to traditional SEO with no platform distinction.

Question 4: Describe a specific technical audit you've performed for an AEO engagement. What did you find, and what did you change?

Score	What It Looks Like
5	Provides detailed, anonymized case study: starting state, audit findings (e.g., missing speakable markup, content structure issues, crawl depth problems), specific changes made, and measurable outcomes (e.g., "citation rate in Perplexity increased from 2 to 11 branded mentions per week over 90 days").
4	Solid case study but light on measurable outcomes.
3	Generic description without specifics.
2	Describes an SEO audit, not an AEO audit.
1	No case study available.

Question 5: How do you handle content being cited inaccurately or out of context by AI engines?

Score	What It Looks Like
5	Explains a defined process: detection (monitoring tools + manual checks), triage (severity assessment), remediation (content restructuring, structured data updates, direct feedback to AI platforms where available), and documentation.
4	Has a process but hasn't formalized it.
3	Acknowledges the problem but offers no concrete remediation approach.
2	Says "we'd update the content" without addressing the AI citation itself.
1	Doesn't recognize this as a real issue.

Question 6: What is your approach to entity optimization and knowledge graph presence?

Score	What It Looks Like
5	Describes Wikidata editing, Google Knowledge Panel management, entity disambiguation strategies, and how they connect entity work to AI answer citation likelihood. Names specific tools (e.g., Kalicube Pro, manual Wikidata contributions).
4	Covers Knowledge Panel optimization but light on Wikidata/entity layer.
3	Mentions "entity SEO" as a buzzword without process detail.
2	Confuses entity optimization with local SEO (Google Business Profile).
1	No answer.

Section 2: Tracking and Reporting

Five questions focused on data ownership, reporting cadence, and transparency. Weighted at 20%.

Question 7: Who owns the tracking data, dashboards, and custom tooling created during the engagement?

Score	What It Looks Like
5	Explicitly states the client owns all data and dashboards. Provides data export in standard formats (CSV, API access). Custom tooling ownership is defined in the contract with clear terms.
4	Client owns data, but dashboards are on agency-hosted platforms (e.g., Looker Studio templates tied to agency accounts).
3	Vague on ownership—"we'll share access."
2	Data lives in proprietary tools with no export capability.
1	Doesn't address data ownership at all.

This mirrors what we flagged in the Website Redesign RFP Template about analytics migration. If the agency walks, your data shouldn't walk with them.

Question 8: What metrics do you report on, and how frequently?

Score	What It Looks Like
5	Reports weekly on citation counts by platform, branded mention accuracy, schema validation status, and content performance against AI-answer triggers. Monthly strategic reviews with trend analysis.
4	Monthly reporting with AEO-specific metrics. Weekly check-ins without formal reports.
3	Monthly reports mixing AEO metrics with generic SEO metrics without clear separation.
2	Quarterly reporting only.
1	"We report when there's something to report."

Question 9: How do you attribute revenue or pipeline impact to AEO efforts specifically?

Score	What It Looks Like
5	Describes multi-touch attribution model that isolates AI-referred traffic (e.g., referral tracking from Perplexity, UTM parameters for ChatGPT Browse clicks, Google AI Overview click-through tracking via Search Console API). Acknowledges limitations honestly.
4	Has attribution methodology but admits it's imperfect. Tracks what's trackable.
3	Claims "we track everything" without explaining how.
2	Falls back on general organic traffic metrics as proxy.
1	No attribution model.

Question 10: Can you provide a sample report from a current or past AEO engagement (anonymized)?

Score	What It Looks Like
5	Provides real, anonymized report including AEO-specific metrics, commentary, and next-step recommendations. Report is clear enough that a non-technical stakeholder could understand it.
4	Provides a sample but it's a template, not a real report.
3	Says they can share one after an NDA is signed (reasonable but delays evaluation).
2	Sample report is clearly an SEO report with "AEO" in the title.
1	No sample available.

Question 11: How do you handle reporting when AI platforms change their citation behavior (as happened with Google's AI Overviews rollback in Q4 2024)?

Score	What It Looks Like
5	Describes specific instance where they adapted reporting methodology to platform changes. Has documented process for recalibrating baselines when AI platform behavior shifts.
4	Acknowledges the volatility and describes general approach to recalibration.
3	Says "we stay on top of changes" without process detail.
2	Unaware of specific platform changes.
1	No answer.

Section 3: Content Production

Six questions about how the agency creates, controls, and validates content for AI answer optimization. Weighted at 25%.

Question 12: Describe your content production model. Do you use in-house writers, contractors, or AI-assisted workflows?

Score	What It Looks Like
5	Transparent about exact mix (e.g., "In-house editorial lead + specialist freelancers for technical verticals + AI drafting with human editing at 60/40 ratio"). Names tools used (e.g., Claude 3.5 for first drafts, Originality.ai for detection, Grammarly Business for style enforcement).
4	Clear model but less specificity on tooling.
3	"We have a team of experienced writers." No detail on AI usage.
2	Evasive about AI usage in content production.
1	No production model described.

Question 13: How do you maintain our brand voice and terminology standards across AEO content?

Score	What It Looks Like
5	Describes voice onboarding process: brand voice audit, creation of style guide or adoption of yours, implementation in AI prompting templates, editorial review checkpoints. References specific quality gates (e.g., "every piece passes through brand voice checklist of 12 items before client review").
4	Has a process but it's less formalized.
3	Says "we match your existing tone" without explaining how.
2	One-size-fits-all content approach.
1	Doesn't address voice control.

Question 14: What is your fact-checking process for content designed to be cited by AI engines?

This question matters more in AEO than traditional SEO. When an AI engine cites your content as factual source, inaccuracies compound—they don't just live on your blog, they get repeated in AI answers to thousands of users.

Score	What It Looks Like
5	Multi-step verification: original source linking (academic papers, primary data), subject matter expert review for technical claims, automated claim-checking tools (e.g., ClaimBuster, manual cross-referencing), and documented correction policy when errors are found post-publication.
4	SME review + source linking, but no automated claim-checking.
3	"Our editors review for accuracy." No described methodology.
2	Fact-checking is the client's responsibility in their model.
1	No fact-checking process.

Question 15: How do you structure content specifically to increase likelihood of AI citation?

Score	What It Looks Like
5	Describes specific formatting patterns: definition-first paragraphs, question-as-heading structures, concise answer blocks under 50 words (matching LLM extraction patterns), proper use of lists and tables for comparison content, and schema markup reinforcing content structure. References testing data.
4	Knows formatting principles, less evidence of testing.
3	"We write clear, authoritative content."
2	Standard SEO content optimization (keyword density, word count targets).
1	No AEO-specific content structuring approach.

Question 16: What is your content update and refresh cadence for AEO-targeted pages?

Score	What It Looks Like
5	Defines refresh schedule triggered by both calendar cadence (e.g., quarterly reviews) and signal-based triggers (citation drop-off, competitor displacement in AI answers, factual obsolescence). Tracks freshness signals per page.
4	Regular refresh schedule but less signal-driven.
3	"We update content as needed."
2	Content is created and not revisited.
1	No refresh strategy.

Question 17: How many content pieces per month does your typical AEO engagement produce, and what types?

Score	What It Looks Like
5	Gives range based on engagement size (e.g., "For $8K/month retainer, typically 6-10 pieces: 2 long-form authority articles, 4-6 answer-format pieces targeting specific AI queries, and 1-2 schema-enriched FAQ page updates"). Types are clearly differentiated.
4	Provides volume and types but less pricing context.
3	"It depends on the scope." No benchmarks.
2	Volume-first approach with no type differentiation.
1	Can't provide volume expectations.

Section 4: Pricing and Commercials

Four questions. Weighted at 15%. As we've noted in our Software Development RFP Template, pricing transparency upfront prevents scope disputes later.

Question 18: What is your pricing model—retainer, project-based, or performance-based—and what does each include?

Score	What It Looks Like
5	Offers multiple models with clear scope definitions. Example: "Retainer: $6K-$15K/month includes X hours technical work, Y content pieces, Z hours strategic advisory. Project-based: $25K-$75K for full AEO audit + 6-month implementation plan. Performance: base + bonus tied to citation growth metrics."
4	One clear model with transparent pricing and scope.
3	"Pricing depends on scope" without any ranges or benchmarks.
2	Only offers one inflexible model.
1	Won't discuss pricing until "discovery call."

Question 19: How do you define and measure success for an AEO engagement, and how do those metrics connect to your pricing?

Score	What It Looks Like
5	Defines 3-5 primary KPIs (e.g., citation count by platform, branded mention accuracy rate, AI-referred traffic, share of voice in AI answers for target queries) with baseline measurement, target-setting methodology, and clear connection to pricing tiers or performance bonuses.
4	Clear KPIs but weaker connection to pricing.
3	Uses traditional SEO metrics (rankings, organic traffic) as AEO success measures.
2	Success is defined vaguely ("increased visibility").
1	No success metrics defined.

Question 20: What are your contract terms, minimum commitment, and exit clauses?

Score	What It Looks Like
5	Minimum commitment of 3-6 months (reasonable for AEO to show results), 30-day written notice to exit after minimum term, no penalty for exit, clear deliverable handoff process, and data portability guaranteed. IP transfer terms are explicit.
4	Fair terms but some ambiguity in exit process.
3	12-month lock-in with early termination fees.
2	Unclear exit terms—"we'll discuss if it comes up."
1	No exit clause or punitive termination terms.

Question 21: Are there additional costs beyond the quoted price (tools, third-party subscriptions, paid placements)?

Score	What It Looks Like
5	Itemized list of what's included vs. what's additional. Example: "Our retainer includes Otterly.AI ($199/mo), our proprietary monitoring stack, and all tool costs. Client is responsible for: any paid content distribution ($500-$2K/mo recommended), subject matter expert time for interviews (estimate 2 hrs/mo)."
4	Most costs disclosed upfront, minor ambiguity on edge cases.
3	"There may be additional tool costs" without specifics.
2	Hidden costs surface only after engagement starts.
1	No cost transparency.

Section 5: Team and Delivery

Three questions. Weighted at 10%. These are the questions most people forget to ask until they're three months in and realize their "senior strategist" is actually a junior account manager forwarding emails.

Question 22: Who is the named delivery lead for our account, and what is their direct AEO experience?

Score	What It Looks Like
5	Names specific person, provides their LinkedIn profile or bio, details their AEO-specific experience (years, client types, results), and confirms they will be day-to-day contact—not just figurehead on pitch deck.
4	Names a lead with relevant experience but can't guarantee long-term assignment.
3	"You'll be assigned a dedicated account manager" without naming anyone.
2	Team is described generically ("our senior strategists").
1	No information about who will do the work.

Question 23: What are your SLAs for response time, deliverable turnaround, and issue escalation?

Score	What It Looks Like
5	Written SLAs: 4-hour response during business hours, 24-hour for non-urgent queries, defined turnaround times per deliverable type (e.g., content pieces: 5 business days, technical fixes: 2 business days, emergency citation issues: same day). Escalation path: account lead → director → principal within 48 hours.
4	SLAs exist but aren't as granular.
3	"We're very responsive" with no commitments.
2	No SLAs—"we handle things as they come."
1	Doesn't understand why SLAs would apply to agency engagement.

Question 24: What happens if our delivery lead leaves your agency or is reassigned?

Score	What It Looks Like
5	Documented transition plan: 2-week overlap period, knowledge transfer documentation, client approval of replacement before transition, option to exit contract if replacement is unsatisfactory.
4	Has transition process but less formal.
3	"We'd assign someone equally qualified." No process detail.
2	No transition plan.
1	Dismisses the concern.

Scoring Methodology

Once you've scored all 24 questions, calculate weighted totals per section:

Section	Questions	Max Raw Score	Weight	Max Weighted Score
Technical Depth	1-6	30	30%	9.0
Tracking + Reporting	7-11	25	20%	5.0
Content Production	12-17	30	25%	7.5
Pricing + Commercials	18-21	20	15%	3.0
Team + Delivery	22-24	15	10%	1.5
Total	24	120	100%	26.0

Scoring interpretation:

22-26: Strong candidate. Move to reference checks and contract negotiation.
17-21: Promising but has gaps. Request follow-up on weak areas before deciding.
12-16: Significant concerns. Only proceed if the market is thin and gaps are addressable.
Below 12: Pass. The agency isn't ready for AEO engagements.

We use the same weighted-section approach in our Website RFP Template—it forces you to make trade-off decisions explicit rather than going with gut feel after reading five proposals back-to-back.

How Does This Template Differ From a Standard SEO RFP?

An SEO RFP evaluates an agency's ability to get your pages ranked. An AEO RFP evaluates their ability to get your content cited as source of truth by machines generating answers.

Key differences:

Schema expertise is non-negotiable in AEO but often nice-to-have in SEO RFPs
Citation provenance tracking has no SEO equivalent—this is entirely new infrastructure
Multi-platform coverage matters because unlike SEO (where Google dominates at ~90% share), AI answers are split across 5+ platforms with meaningful market share
Fact-checking standards are higher because AI amplifies errors at scale
Attribution modeling is harder and any agency claiming clean attribution is lying or selling something

What Should You Include in the Background Section of Your RFP?

Before the 24 questions, your AEO agency RFP needs company background. Include:

Your industry and primary audience—be specific ("B2B fintech serving CFOs at companies with $50M-$500M revenue" not "financial services")
Current AEO baseline—are you being cited anywhere today? Do you know? If not, say so.
Existing content infrastructure—CMS, current content volume, publishing cadence, team size
Budget range—give a range, not single number. $5K-$8K/month signals very different things than $15K-$25K/month. Agencies calibrate their proposals to budget.
Timeline expectations—when you need initial results (hint: anything under 90 days for AEO is unrealistic)
Decision-making process—who evaluates proposals, who signs contract, what's the approval chain

Altitude Marketing's RFP guidance correctly emphasizes that agencies need four key dates working backward from completion. We'd add a fifth for AEO: the date you'll re-evaluate the engagement's ROI model, because AEO attribution is still maturing and your initial metrics will likely need recalibration at 90 days.

FAQ

How many agencies should I send this AEO RFP to?

Send to 3-5 agencies maximum. The AEO agency market is small enough that sending to more creates evaluation overhead without improving your odds. Focus on agencies that have published AEO-specific case studies or thought leadership, not agencies that added "AEO" to their services page last month.

What budget should I expect for AEO agency services in 2025-2026?

Retainers typically range from $5,000-$20,000 per month depending on scope, content volume, and number of AI platforms covered. Project-based AEO audits run $15,000-$50,000. Performance-based models are rare because attribution is still immature. Be skeptical of anything under $4,000/month.

Can I use this template alongside a traditional SEO RFP?

Yes, and we'd recommend it if you're evaluating agencies claiming to do both. Send the SEO-specific questions from our Website Redesign RFP Template alongside this AEO template. How an agency differentiates their answers between the two tells you a lot about actual depth.

What's the biggest red flag in an AEO agency proposal?

An agency that can't name specific AI platforms they optimize for, or that treats AEO as "just SEO with better structured data." Also watch for agencies promising specific citation counts—no one controls what an LLM decides to cite. Process quality is what you're buying.

Should I require a pilot period before committing to a full AEO engagement?

Absolutely. A 90-day pilot with defined scope (e.g., optimize 10-15 pages for AI citation, track results across 3 platforms) at 40-60% of full retainer cost is reasonable. Any agency refusing a pilot is either overconfident or worried the results won't hold up.

How do I evaluate AEO agencies if I have no internal AEO expertise?

Use this scoring rubric as-is—it's designed so you can score based on specificity and evidence without needing to be an AEO expert yourself. Agencies scoring 4-5 will naturally provide enough detail for you to verify their claims. Agencies scoring 1-2 will be vague enough that lack of expertise is self-evident.

Download the full .docx template | Browse all RFP templates