Action gap
Dashboards tell you where you appear. Not what to do next.
A visibility score is not a strategy. We needed an instrument that names what to do next, on real data.
Most AI visibility trackers tell you where you're invisible.Ours tells us what to do about it.
We use our visibility tracker to test buyer prompts (grounded in your data) across major AI engines to understand where and how you're showing up. You get a data advantage from our work across B2B SaaS companies, we get clarity.
Tracking AI visibility across
- ChatGPT
- Claude
- Perplexity
- Google AIO
- Gemini
Performance
ExportCitation Rate
42.8%
▲ +6.2%Prev: 36.6%
Mention Rate
61.4%
▲ +4.1%Prev: 57.3%
Average Rank
#2.4
▼ −0.8Prev: #3.2
Share of Voice
34.2%
▲ +7.8%Prev: 26.4%
40%30%20%10%0%
Jan 06Jan 27Feb 17Mar 10Mar 31
Competitive ranking
1.HubSpot26.4%34.2%+7.8
2.Salesforce28.1%22.4%−5.7
3.
Pipedrive11.8%14.0%+2.2
Pipedrive11.8%14.0%+2.24.
Attio8.2%12.8%+4.6
Attio8.2%12.8%+4.65.Zoho CRM9.1%9.4%+0.3
6.
Close5.4%4.8%−0.6
Close5.4%4.8%−0.67.
Copper2.4%2.4%—
Copper2.4%2.4%—Topical Performance
Pillar / Cluster
HubSpot
SFDC
PipedriveAttioZoho
Sales pipeline mgmt
42.1%
21.4%
14.8%
9.2%
6.4%
Lead management
38.6%
22.8%
11.4%
8.6%
5.2%
Sales automation
31.2%
34.6%
12.4%
6.8%
5.4%
Reporting & analytics
39.4%
24.2%
8.4%
11.6%
7.2%
Providing B2B companies with a data advantage.
Most AI visibility tools give a dashboard.None of them provide a plan.
We tried the tools. We didn't like them so we built our own. Clients get view-only access while we run the strategy.
Randomness gap
Most tools mistake the noise for movement.
LLMs answer the same question differently every time. Most tools don't show statistically significant results. We use the math to find what is real.
Perception gap
A mention is not always a win.
Being positioned badly can hurt more than not being mentioned at all. We score answers for accuracy and depth, then fix how you show up.
Prompt gap
Generic prompts give generic answers.
We ground prompts in your call transcripts, win and loss notes, and real buyer language. No leading questions. No flattering prompts.
The technology behind real B2B SaaS pipeline.The system our team uses every day.
Oldschool agencies still operate from spreadsheets. We're building a new operating model that gives richer insights and faster results.
+22%
organic meetings booked at incident.io
167%+
increase in organic demos at Sova
Powered by CITABLE
Seven signals that decide what AI cites. Tested against 1M+ live answers a month.
99% of agencies suck. But you guys do not. One month in, our referrals from ChatGPT are up 29%. And we can finally see why.
.jpeg)
Dan Harvey
Co-founder, Passion.io
“There are orgs like HubSpot and Ramp with dedicated AEO teams. For everyone else (except my competitors), there's Discovered Labs.”
Tom Wentworth
CMO, incident.io
“You guys make sure our name pops up whenever someone is thinking about talent assessment. The tracker shows exactly when, and where.”
Sabina Reghellin
Content Marketing Manager, Sova
How we use our AI visibility trackerto get you better results
Most AEO dashboards report rate moves without bounding the noise. Our tracker has to clear three checks before we call a movement real to avoid being fooled by randomness.
/ 01Variance bounds
Every number carries a 95% credible interval.
We test each buyer prompt multiple times to bound the noise. The credible interval shows us how confident to be in any movement before we act on it.
/ 02Prompt-set hygiene
Prompts pass coverage, confound, and brand-anchor checks.
Sampled from your GSC, support, and Reddit signal. Not hand-picked. We strip leading prompts that artificially inflate mention rate.
/ 03Trend significance
A move counts only if the math is decisive across three consecutive runs.
The change has to clear a strict statistical bar, and it has to hold week over week. No false-positive feeds.
Methodology:Noise vs. signal in AEO
Know who you actually compete against in LLMs.
The competitor set LLMs serve in your category is rarely the one you assume. We configure it with you in onboarding then track share of voice per engine, week by week.
- Benchmark against your competitive set
- Share of voice over time, per engine
- Segment by platform
Competitor trackingBenchmark performanceTrend over time
Competitive rankingPipedriveAttioClose
40%20%0%
HubSpot
Mar 02Mar 16Mar 30
#
Brand
Now
Δ
1
HubSpot
34.2%
+7.8
2
Salesforce
22.4%
−5.7
3
Pipedrive14.0%
+2.2
4
Attio12.8%
+4.6
5
Close4.8%
−0.6
Find the topics you are winning and losing.
Drill from pillar to topic to individual prompt to see where you show up and where competitors lead. The next strategy decision becomes data-led, not a hunch.
- Drilldown by topics you want to own
- Topical share of voice per competitor
- See performance by prompt
Topical performancePrompt performanceUnderstand drivers
Topical performancePipedriveAttio
Pillar / Topic
HubSpot
SFDC
PipedriveAttioZoho
Sales pipeline mgmt
42.1%
21.4%
14.8%
9.2%
6.4%
Lead management
38.6%
22.8%
11.4%
8.6%
5.2%
Sales automation
31.2%
34.6%
12.4%
6.8%
5.4%
Reporting & analytics
39.4%
24.2%
8.4%
11.6%
7.2%
Service & support
28.4%
19.6%
8.2%
5.4%
4.8%
Marketing automation
33.8%
27.1%
9.6%
7.2%
5.0%
Customer success
36.2%
18.4%
10.1%
6.6%
4.2%
See which Reddit threads, reviews, and articles LLMs actually trust.
We identify the subreddits, threads, and third-party sources LLMs cite when answering questions in your category. The Reddit, PR, and editorial plan we implement is informed by what is moving AI answers, not just by what ranks in Google.
- Understand citation trends
- Subreddits and threads driving citations
- Inform off-page activities
Earned mediaReddit marketingDigital PRLink building
Cited sources
Breakdown by source type
Reddit29%
Reviews22%
YouTube17%
Blogs14%
Owned10%
Other8%
#
Source
Citations
Status
01
reddit.com/r/CRM
418
Covered
02
g2.com/categories/crm
302
Queued
03
producthunt.com
214
Covered
04
salesforce.com/blog · crm-comparison
187
Gap
05
capterra.com/crm-software
152
Queued
06
youtube.com — SaaStr podcast
128
Covered
See how LLMs perceive you and your competitors.
We score every LLM response across multiple dimensions including features, pricing, differentiators, and funnel stage. Drift in how LLMs describe you surfaces before it influences a buyer's shortlist.
- Brand perception across multiple dimensions
- Side-by-side scoring versus every tracked competitor
- Recency, accuracy, and detail graded on every response
Brand perceptionCompetitor scoringPositioning drift
Brand perceptionPipedriveAttio
Category
HubSpotYOU
Salesforce
PipedriveAttioGap
Awareness
81
73
77
68
+4
Competitive
79
64
60
69
+10
Discovery
44
61
30
11
−17
Evaluation
72
66
70
69
+2
Features
68
63
66
61
+2
What lives inside the tracker
Each feature is a slice of what our team uses every day to get you results.
Prompt cohort tracking
Real buyer prompts (not hand-picked) tested every 5 days across every model. We see when LLMs name you and when they pick a competitor.
Recommend a CRM for a 12-person sales team. Budget is tight and we need something up and running this week.
For a small team on a tight budget, HubSpot's free tier is the fastest way to get started.
Ask AI Assistant
+
Multi-model coverage
Same prompt across ChatGPT, Claude, Perplexity, Copilot, and Google AI Overviews. Locale-aware. No blind spots.
Copilot
ChatGPT
ChatGPTCited
Claude
PerplexityCited
CopilotCited
ChatGPT
Citation source mix
Reddit, reviews, YouTube, blogs. We see which sources LLMs trust in your category, ranked by citation volume.
Reddit29%
Reviews22%
YouTube17%
Blogs14%
Owned10%
Other8%
Share-of-voice trend
Your share of voice tracked over time, with the gap to category leaders visible at a glance.
34.2%
▲ +7.8
Share of voice · 30d
Brand perception scoring
We score how LLMs describe you across multiple dimensions including features, pricing, differentiators, and funnel stage. Head-to-head against every tracked competitor.
Radar alerts
Get pinged the moment a competitor publishes new content, refreshes an old page, or breaks into a listicle you weren't in.
Salesforce published new comparison page
2h ago
Pipedrive refreshed pricing page (you: 14mo old)
5h ago
Attio added to “Best CRMs for startups” on G2
Yesterday
Frequently askedquestions.
Still curious? Book a call and we'll walk through it on your brand.
How is the AI Visibility Tracker different from Profound, Peec.ai, or Otterly?
Profound, Peec.ai, and Otterly are SaaS dashboards you operate yourself. The Discovered Labs tracker is the instrument behind a service. We run the prompts, score the answers, separate signal from noise, then our team of senior marketers builds and implements your organic search strategy. We're accountable for performance, DIY software is not.
Can I use the tracker without the Discovered Labs retainer?
No. We're not a SaaS company. The tracker exists to power the strategy we run for you, so it isn't sold as standalone software. Every retainer client gets view-only access so you can see the data we're working from in real time.
Why use an agency instead of buying an AEO tracking tool ourselves?
Because the tool isn't the bottleneck. The bottleneck is what to do once you have the data. We do that work end to end. You get the tracker as the instrument and a team that operates it as the differentiator.
AI answers change every time you ask. How do you separate signal from noise?
LLMs answer the same question differently every time, so a single run is noise. We run each buyer prompt multiple times, then bound the result with a 95% credible interval. A movement only counts if it clears three checks: the credible interval has to move, the change has to be statistically decisive, and it has to hold across three consecutive runs.
Which AI engines does the tracker cover?
Five engines by default: ChatGPT, Claude, Perplexity, Google AI Overviews, and Gemini. Each prompt runs across every engine on the same cadence so you can compare share of voice engine by engine and spot drift in one before it spreads to the others.
What's included in a Discovered Labs retainer?
A retainer covers the full search and AI search operation: content using our CITABLE framework, landing pages, original research studies, technical SEO and AEO audits, as well as off-page activities such as Reddit and link building. The visibility tracker is included as part of the operation. View our pricing page for up-to-date information on packages.
Can't we just use Semrush AI Visibility or Ahrefs Brand Radar?
You can. They will give you a directional read on whether your brand shows up in AI answers, and that's useful if you only need a number. Some of our clients also keep their existing subscriptions.
How do you decide which prompts and competitors to track for us?
We configure both with you during onboarding, then frequently revisit. Competitors come from you during onboarding as well as our own research and what companies you appear next to inside LLMs.
Prompts are sampled from your GSC queries, support tickets, customer call transcripts, and the subreddits your category lives in. Every prompt then clears three hygiene checks before it enters the bench: coverage (is it a real buyer question), confound (does it accidentally trigger an unrelated category), and brand-anchor (does it lean toward you or a competitor).
Ready when you are
Ready to improve performance of search?Book a call with us and we'll walk you through our methodology live.
30 minutes with our team. We'll walk you through where AI search has your brand today, the topics your buyers are asking about, and what to do about it. Across AI assistants and Google. Whether you work with us or not.