GEO Claims and Evidence
Assess GEO Claims Before You Sign
Check outcome promises, reporting evidence and the scope behind a quote. A practical guide to evaluating GEO claims without relying on fear, anonymous allegations or incomparable prices.
v1.1 · October 2026 · Tenten GEO Research
CH.01
Start with a Claim You Can Check
A procurement decision needs evidence about the proposed work. An alarming story about another company is not a substitute for a defined scope, measurement method and responsible team.
Before relying on a case study, identify the reporting source, observation period, exact outcome and measurement method. Confirm that the material may be shared. An anecdote without those details can suggest questions, but it cannot establish a supplier’s effectiveness or misconduct.
| Claim to examine | Evidence to request | What it can establish |
|---|---|---|
| Who is responsible? | A verifiable business identity, named roles and relevant experience | Accountability; company age or a headshot alone cannot establish competence |
| What outcome is promised? | The exact metric, conditions, period and remedy | Whether the offer concerns delivery, a refund or a platform outcome |
| What does the report measure? | Prompts, product interface, dates, sample and denominator | The observation behind a score, with its limits |
| What can be inspected? | A permitted sample, redacted record or clearly labeled demonstration | How the method works without exposing another client’s confidential data |
Tenten’s questions for evaluating a proposal; not findings about a particular vendor
Assess the answer and the evidence together. A new firm, an NDA or a proprietary tool can have legitimate reasons. A missing document is a gap to resolve; it is not, by itself, proof of fraud.
What to do
Write down the claim, how it would be checked and who is responsible. If a material question remains unresolved, narrow the commitment or defer the decision.

CH.02
Three Claims That Need Clarification
Strong language can hide an undefined offer. Ask what the statement means in measurable terms before treating it as evidence for or against a vendor.
Script one
Guaranteed placement
Ask whether the promise means a defined deliverable, a commercial remedy or guaranteed inclusion in a third-party answer. Google states that eligible pages are not guaranteed to be crawled, indexed or served. A refund promise does not give an agency control over that decision.
Script two
A deadline driven by fear
Ask for evidence relevant to your market, existing traffic and buyer needs. “Act now or disappear” does not establish the cost of waiting, and a historic traffic share from another site does not establish your opportunity.
Script three
A proprietary method
The vendor may protect implementation details while explaining inputs, outputs, sampling limits and evaluation. Agree what you can inspect. An unexplained score should not be the only basis for acceptance or payment.
Separate delivery commitments from predictions. Ask for the assumptions behind a forecast and the evidence needed to revise it. Do not turn a confident pitch into an assumed guarantee.
What to do
Restate the offer in plain language. If both sides cannot agree what success, uncertainty and a missed commitment mean, resolve that before expanding the work.
CH.03
Uncertain Outcomes Still Need Accountable Delivery
A platform outcome may be outside a supplier’s control. That does not remove responsibility for the work, evidence and communication the supplier agreed to provide.
Search products, retrieval settings and prompts can change. Specify which product and interface are measured, and keep the collection conditions with the result. An API benchmark should not be presented as a complete record of what every buyer sees.
A before-and-after improvement is an observation. Ask what else changed and whether the study supports a causal claim. A well-run process can produce no measurable lift; the report should explain that plainly, alongside the next decision.
- →Define deliverables: pages, technical fixes, evidence records and responsible reviewers.
- →Define measurement: sample, denominator, missing responses and material limitations.
- →Define the decision point: continue, change scope or stop, based on the agreed evidence.
What to do
Ask for a sample that explains both results and uncertainty. A clear account of a limited result is more useful for evaluation than an unsupported claim of certainty.
CH.04
Evidence You Can Reasonably Inspect
Agree on evidence appropriate to the service and privacy obligations. A demonstration can show how a report works, but it must not be presented as a real customer result.

| Evidence area | Insufficient on its own | Useful context |
|---|---|---|
| Citation tracking | A single headline visibility score | Prompts, platforms, dates, responses, denominators and missing data |
| Content changes | “Content optimized” without a record | The affected page, change, reason and review responsibility |
| Result comparison | An unlabeled trend chart | Comparable periods, unchanged sample definitions and alternative explanations |
| Method explanation | A score with an undisclosed meaning | Inputs, outputs, limits and what can be independently checked |
Tenten’s reporting questions; records can still contain errors and need appropriate review
Demand one
Request an appropriate sample
Ask for a permitted historical example or a clearly labeled demonstration. Redaction, an NDA or an independent reference may be reasonable. Confirm which claims the available evidence actually supports.
Demand two
Agree on the baseline
Define the initial sample before measuring change. A scoped audit can be paid work; ask what it includes, who owns the output and whether it is usable if no later engagement follows.
Demand three
Agree on reporting and handover
Specify the evidence, cadence, access, retention and handover in the proposal. Match effort to the decision rather than asking for repeated checks with no defined purpose.
What to do
Agree what can be shared without exposing another client. If the available evidence is insufficient for the proposed commitment, reduce the scope or seek a different arrangement.
CH.05
Compare the Scope Behind the Price
A monthly advisory fee, a software subscription and an enterprise implementation budget buy different things. Dividing the largest figure by the smallest does not produce a useful quality benchmark.
| Engagement type | Scope to establish | Cost questions |
|---|---|---|
| Advisory or software | Recommendations or measurement access; who implements changes | Seats, prompts, usage, support and internal time |
| Ongoing delivery | Content, technical work, measurement and review responsibilities | Included work, limits, dependencies and additional charges |
| Implementation project | Markets, systems, migration, integrations and handover | One-off fee versus recurring costs; acceptance and maintenance |
A scope comparison, not a survey of current market prices
Stackmatix’s March 2026 pricing article distinguishes US$25,000–100,000 implementation projects from US$10,000–20,000+ monthly enterprise retainers. Its separate enterprise article describes broader total-program budgets. These are vendor-published guidance with different scopes, not comparable quotes for one service or an independent market sample. The former 200× chart is retired.
What to do
Compare currency, billing period, included work and internal effort on the same basis. A low price does not prove poor practice, and a premium price does not prove expertise. For Tenten’s current starting prices, use the Pricing page and the written quote for your scope.
CH.06
A Short Checklist for the Decision
Use these questions to organize unresolved issues. This is an editorial checklist, not a fraud-detection score or a substitute for your procurement process.
| Question | Useful evidence | What to resolve |
|---|---|---|
| Identity | Verifiable company and delivery responsibilities | Who can answer for the work? |
| Outcome | Defined metric, conditions and commercial remedy | What is controlled and what is a forecast? |
| Priority | Evidence relevant to the business problem | Why this work now? |
| Method | Inputs, outputs and measurement limits | What can be inspected or reproduced? |
| Delivery | Scope, owner, records and review process | How will a missed commitment be handled? |
| Baseline | A defined initial sample and collection conditions | How will change be interpreted? |
| Price | Comparable scope and total expected cost | What is excluded or separately charged? |
Seven questions for an informed decision; no numeric pass score is implied
Step one
Review the evidence
Ask the vendor to distinguish customer evidence, demonstrations, estimates and public research. Record material gaps instead of treating all examples as equivalent.
Step two
Agree the initial scope
Choose a commitment appropriate to the evidence, budget and internal capacity. Confirm the baseline and responsibility for implementation and review.
Step three
Confirm the decision process
Put deliverables, access, review points and handover into the agreed documentation. Obtain clarification on commercial terms through the normal procurement process.
The next decision
Choose a provider whose scope and evidence fit your need. If you cannot explain what you are buying and how it will be assessed, clarify the proposal before signing.
Sources & methodology
Updated October 5, 2026. This is Tenten’s editorial procurement guidance, not a finding of fraud or an independent vendor ranking. The previous TheAdSpend-attributed Toronto case could not be matched to an original report in this review and is no longer used as evidence. The former 200× price chart combined incomparable scopes and was removed; vendor context is documented at https://www.stackmatix.com/blog/aeo-services-pricing and https://www.stackmatix.com/blog/enterprise-aeo-services . Google’s eligibility guidance is at https://developers.google.com/search/docs/appearance/ai-features . These sources do not validate every vendor’s outcome or a universal service price. No private client records or real performance results were obtained. Commercial commitments are governed by the agreed proposal and terms.
© 2026 Tenten GEO · Feel free to cite this whitepaper with attribution.
Discuss your AI search priorities
Request a 30-minute benchmark interpretationSend a booking request. We’ll email you to confirm a time within 24 hours.