Share of voice
Compare your mention count with the mentions of tracked competitors in the recorded answer set for the selected period.
AI Competitor Tracking
See who AI recommends before your brand—and where they win
Compare mentions, order of appearance, sentiment, and share of voice across the same buyer questions, then open the exact answers behind every competitive gap.
Competitive metrics cover the prompts and API answers recorded inside your project. They are not market share estimates or consumer-app traffic data.
Product scope
What this page covers
- Tracked set
- Named + discovered
- Configured competitors plus additional brands surfaced in answers
- Comparison
- Engine + topic splits
- See where a lead exists instead of averaging it away
- Audit trail
- Answer-level proof
- Open the transcript behind a mention or competitor win
The real competitive result
A mention is not the same as a recommendation
Two brands can appear in the same answer and occupy very different positions. One may be the first recommendation with a positive explanation; the other may be an afterthought at the bottom of the list. A binary mentioned-or-not report hides the difference that matters to a buyer.
SEO Address records the order in which each brand first appears, the sentiment attached to the mention, and the topic and engine that produced it. Those observations roll into a visibility score and share of voice, while the original response remains available whenever a number needs interpretation.
Competitive visibility is most useful when you can move from a portfolio metric to the specific topic, prompt, engine, and wording behind it.
How it works
Compare the same buying situations—not unrelated anecdotes
A controlled project gives every tracked brand the same opportunity to appear in the same prompt set.
- 1
Name the brands in the decision set
Add your brand and the competitors you want to compare, including their domains. The analysis also retains untracked brands it discovers, so the view can reveal a challenger you did not expect.
Output: A defined competitive set with room for discovery
- 2
Run shared prompts across both supported providers
Every brand is measured inside the same OpenAI and Anthropic answers. Topic labels make it possible to distinguish broad category strength from a narrow use-case win.
Output: Comparable evidence by prompt, topic, and engine
- 3
Measure frequency and prominence
Count mentions, first-place appearances, and reciprocal rank. Share of voice shows your share of tracked-brand mentions; visibility rewards earlier placement across all answers.
Output: Metrics that separate presence from prominence
- 4
Open the gap before choosing an action
Review the response where a competitor appeared and you did not, then examine its cited domains and wording before deciding whether content, positioning, or outreach is the right response.
Output: A grounded competitive diagnosis
Competitive intelligence
See the lead from more than one angle
Each view answers a different question, and each can be traced back to recorded answers.
Visibility score
Reward early appearances using reciprocal rank across all recorded answers, so first position carries more weight than a late mention.
First-place and mention counts
Separate how often a brand appears from how often it is the first detected brand in the response.
Mention sentiment
Classify a mention as positive, neutral, or negative and retain a supporting quote to make review practical.
Topic-level gaps
Find buying situations where competitors are present and your brand is absent, rather than treating a category-wide average as the answer.
Discovered alternatives
Retain brands beyond the configured list and surface the highest-volume discovered names alongside your tracked set.
Metric definitions
Know exactly what each competitive number means
The dashboard uses observable signals from recorded answers rather than an opaque estimate of model preference.
Metrics are calculated from successful answers in your configured monitoring set. Changing the prompt set changes the population being measured.
Product record
- Visibility score
Reciprocal rank across answers
Earlier brand appearances contribute more; the result is capped at 100
- Share of voice
Your mentions ÷ tracked-brand mentions
A comparison inside the recorded set, not total market share
- Mention rate
Answers mentioning you ÷ all answers
Shows breadth of presence across monitored responses
- Competitor win
Competitor named while you are absent
The report links to an example answer for context
Questions this page answers
Turn a vague competitive concern into a specific investigation
01
Who owns the shortlist?
See which tracked and discovered brands appear most often and earliest across the buyer questions you monitor.
02
Where is the lead concentrated?
Break results down by engine and topic to distinguish a broad visibility problem from one narrow content gap.
03
What changed after publishing?
Compare repeated observations over time, then inspect the raw answers before connecting a movement to a campaign or page update.
Read competitive AI metrics with the right boundaries
The numbers describe a monitored sample of generated answers. They should guide investigation, not be presented as universal market measurements.
- Share of voice is your share of tracked-brand mentions in recorded answers. It is not revenue share, traffic share, or the share of every AI conversation in the market.
- Results depend on the prompts, topics, engines, competitors, aliases, and cadence configured in the project. Material changes to that setup can break a clean trend comparison.
- OpenAI and Anthropic API answers vary from run to run and do not exactly reproduce personalized consumer ChatGPT or Claude sessions.
- The product explains where a competitor appeared and what sources were returned; it does not reveal a model's private training data or prove why the model chose a brand.
Common questions
Questions about ai competitor tracking
Clear answers about scope, evidence, and what the current product actually does.
What does AI competitor tracking measure?
It measures how your brand and configured competitors appear across the AI answers recorded for your project: mention count, first appearance order, first-place appearances, sentiment, share of voice, visibility score, and performance by topic and provider.
How many competitors can I track?
The current project model accepts up to 10 configured competitors as a technical cap, not a pricing promise. The analysis can also retain other brands detected in answers and surface high-volume discovered alternatives that were not in your original list.
How is AI share of voice calculated?
SEO Address divides your brand's mentions by the total mentions for your brand and the configured competitors in the recorded answer set. It is a project-level comparison, not an estimate of all conversations happening inside consumer AI products.
Why does order of appearance matter?
A brand introduced first is often more prominent than one added late in a long list. The visibility score therefore uses reciprocal rank: first place contributes 1, second place 1/2, third place 1/3, and so on, before the result is normalized across answers.
Can I see where a competitor beat me?
Yes. Topic views show where competitors have more mentions, and weekly reports retain example answer IDs for competitor wins. You can open the underlying answer to review the prompt, wording, provider, mentions, and returned citations.
Does a higher score prove that a competitor will win more customers?
No. The score describes prominence in your monitored AI answer set. It can identify a discovery and positioning risk, but it does not measure conversions, revenue, or the behavior of every buyer who uses an AI assistant.
Continue with the next useful page
AI prompt monitoring
Build the consistent buyer-question set that makes a competitive comparison meaningful over time.
Open pageWeekly AI visibility reporting
Summarize week-over-week movement, competitor wins, and the gaps that need attention next.
Open pageAI visibility metrics explained
Learn when to use mentions, position, share of voice, sentiment, and citations—and what each metric leaves out.
Open pageMeasure the shortlist your buyers may be seeing
Add your competitive set, approve the questions, and keep the evidence behind every AI visibility comparison.
Start competitor tracking