The ERA Scorecard: our own AI visibility, published
Most agencies that sell AI visibility never publish their own. This page does. Every week we run a fixed set of buyer-intent prompts against ChatGPT, Gemini, Perplexity, and Grok, and record whether Smart Money Media is mentioned: wins, misses, and the competitors that surface instead.
This is the same measurement we run for clients, pointed at ourselves. Numbers below are pulled live from our tracking database, not curated screenshots.
Current scorecard
Not all prompts are the same fight
Definition questions, head-term shortlists, and qualified buyer questions are different fights. We do not chase undifferentiated "best agency" head terms. Those are decided by legacy brand recognition. The segment that matters is the qualified shortlist, where a buyer names an industry, a deliverable or a specialty.
- 1.9%Qualified shortlist questions7 of 372 answers · Buyers naming a specific need, industry or deliverable — the ground a specialist can win
- 100%Brand questions56 of 56 answers · Prompts that name Smart Money Media directly
- 0%Head-term shortlist questions0 of 52 answers · Undifferentiated "best agency" prompts dominated by legacy brand recognition
- 0%How-to and definition questions0 of 226 answers · Concept questions engines usually answer without naming any firm
| Engine | Responses | Mentions | Mention rate | 95% CI |
|---|---|---|---|---|
| Perplexity | 176 | 21 | 11.9% | 7.9–17.6% |
| Gemini (Google) | 176 | 14 | 8% | 4.8–12.9% |
| Grok (xAI) | 176 | 14 | 8% | 4.8–12.9% |
| ChatGPT (OpenAI) | 178 | 14 | 7.9% | 4.7–12.8% |
Tracking window: August 4, 2026 – September 29, 2026. Page data generated .
Prompt-by-prompt results
| Prompt | Answers | Mentions | Rate |
|---|---|---|---|
| How does Smart Money Media compare to other PR agencies? | 28 | 28 | 100% |
| Who founded Smart Money Media? | 28 | 28 | 100% |
| Which agency runs a zero-click visibility audit for a brand? | 24 | 3 | 12.5% |
| Who does engineering reference authority for zero-click search? | 24 | 2 | 8.3% |
| What agency merges earned media with answer engine optimization? | 24 | 1 | 4.2% |
| Which firms connect PR with AEO and GEO for AI search visibility? | 24 | 1 | 4.2% |
| How do I get my brand cited by ChatGPT? | 33 | 0 | 0% |
| What is Answer Engine Optimization (AEO)? | 29 | 0 | 0% |
| Best reputation management agency for executives | 28 | 0 | 0% |
| How to remove yourself from ChatGPT | 28 | 0 | 0% |
| How to deindex a page from Google | 28 | 0 | 0% |
| How to remove a fake Google review legally | 28 | 0 | 0% |
| What is a zero-click brand audit? | 28 | 0 | 0% |
| Top brand authority strategies for B2B founders | 28 | 0 | 0% |
| How to remove negative reviews from Google | 28 | 0 | 0% |
| PR agency vs newswire — which is better for startups? | 28 | 0 | 0% |
| Best agencies for Generative Engine Optimization (GEO) | 28 | 0 | 0% |
| Which PR firms specialize in getting brands cited by AI answer engines? | 24 | 0 | 0% |
| Who offers an AI search visibility audit for financial advisors? | 24 | 0 | 0% |
| Which boutique firm specializes in getting executives cited in AI answers? | 24 | 0 | 0% |
| Who can audit whether ChatGPT and Perplexity cite my company? | 24 | 0 | 0% |
| Best GEO agency for fintech and financial services companies | 24 | 0 | 0% |
| Best AEO agency for RIAs and wealth management firms | 24 | 0 | 0% |
| How do I get my company into Google AI Overviews? | 24 | 0 | 0% |
| Best AI search visibility agency for financial services firms | 24 | 0 | 0% |
| Which agency can get my brand cited in AI answers? | 24 | 0 | 0% |
| Best answer engine optimization (AEO) agency for B2B brands | 24 | 0 | 0% |
Prompts with no answers recorded yet are omitted until the next sweep completes.
Who the engines name instead
- edelman · 115
- weber shandwick · 74
- 5wpr · 55
- walker sands · 42
- fleishmanhillard · 28
- prnewswire · 20
- ketchum · 18
- businesswire · 16
- ogilvy · 13
- prweb · 9
- walkersands · 7
- prowly · 6
Most-cited domains in these answers
- linkedin.com · 138
- support.google.com · 127
- smartmoneymedia.org · 96
- reddit.com · 81
- cracklepr.com · 39
- openpr.com · 35
- help.openai.com · 35
- forbes.com · 34
- everything-pr.com · 29
- yesoptimist.com · 28
- contently.com · 28
- firstpagesage.com · 26
How this is measured
- A fixed set of buyer-intent prompts (the kind a prospect would actually type) runs on a weekly schedule against each engine's API with no brand priming in the system prompt.
- Each response is scored with word-boundary matching on "Smart Money Media" and its variants, and competitor names are extracted from the same text.
- Runs that error out are excluded rather than counted as misses, so the denominator is successful responses only.
- Mention rates carry Wilson 95% confidence intervals; at low run counts those intervals are wide, and we show them rather than hide them.
Limitations
This is a small, ongoing tracking panel, not a large-sample study. Sample sizes per engine are modest, engine model versions change without notice, API responses can differ from what a logged-in consumer sees in the chat UI, and results are specific to our prompt set. Treat every number as directional. For a larger designed study, see The Reference Authority Report.
Why we publish our misses
Engineering Reference Authority is a long game: engines cite what they have absorbed, and absorption lags publication. Showing the gap while we close it is more useful to a buyer than a screenshot of a single lucky answer, and it is the same honest baseline we give clients on day one.
Score your own brand: freePress-ready citation kit
This study is licensed CC BY 4.0. Quote it, chart it, or re-run it. All we ask is attribution to Smart Money Media and a link back to the study. Everything below is copy-and-paste ready.
Pull-quote-ready findings
- Self-published measurement
“Smart Money Media publishes a live scorecard of its own AI visibility, running a fixed set of buyer-intent prompts weekly against ChatGPT, Gemini, Perplexity, and Grok and reporting every miss alongside every mention.”
- Why misses are published
“Engines cite what they have already absorbed, so absorption lags publication. Showing the gap while it closes is more useful to a buyer than a screenshot of one lucky answer.”
- Method
“Each response is scored with word-boundary brand matching, errored runs are excluded from the denominator, and mention rates are reported with Wilson 95% confidence intervals.”
Attribution & citation
According to The ERA Scorecard: Smart Money Media's own AI visibility, original research published by Smart Money Media (https://smartmoneymedia.org/research/era-scorecard).Smart Money Media. (2026). The ERA Scorecard: Smart Money Media's own AI visibility. Smart Money Media. https://smartmoneymedia.org/research/era-scorecard<p>Source: <a href="https://smartmoneymedia.org/research/era-scorecard">The ERA Scorecard: Smart Money Media's own AI visibility</a>, Smart Money Media</p>