What does adding a competitor do?
Adding a competitor tells Attensira to look for that brand in the answers it is already collecting. No extra runs are issued and no extra prompts are consumed. Every answer from every tracked prompt is re-read for the competitor’s name and domain, and the result becomes a competitor mention rate and citation rate next to your own. Because the analysis runs over answers you already have, a competitor added today is measured against answers collected from that point forward, on the same daily cadence as everything else. Competitors do not consume prompt slots. Add a competitor by domain —example.com — rather than by display name. The domain is what makes a citation attributable; the brand name is what makes a mention detectable. Attensira uses both.
You can add one programmatically with the add_competitor MCP tool. See MCP overview.
What gets measured for a competitor?
The same two rates you get for yourself, over the same runs:- Mention rate — successful runs that name the competitor, divided by all successful runs.
- Citation rate — successful runs that attach a URL on the competitor’s domain, divided by all successful runs.
Why don’t competitor rates add up to 100%?
Because a single answer can name several brands at once, and every rate is measured against the same denominator: all successful runs in the window. Each brand is scored independently — “was this brand named in this answer, yes or no” — so the rates overlap by design and routinely total well over 100%. They can also total under 100%, when many answers name nobody you track. Every competitive product you have used implies the opposite. Share of voice, market share, ranking positions — these all divide a fixed pie and sum to one. Attensira does not, because an assistant answering “what are the best expense tools for a small agency?” does not pick a winner. It lists five.A worked example
Ten successful runs of one prompt. Here is what the answers contained:
That totals 170%. Nothing is wrong. Acme appears in seven of ten answers; you appear in five; they appear together five times. Run 7 named nobody, and it still sits in every denominator — an answer that ignored the whole category is real evidence, not a missing data point.
The practical consequence: your mention rate going up does not require a competitor’s to go down. You can both rise, if models start naming more products per answer. Read each rate as “how often is this brand in the room,” never as “what slice of the room does this brand own.”
The engine has a field named
ShareOfVoice. Despite the name, it holds the brand’s own mention rate — the same number defined above, with the same denominator. It is not a share of anything and it does not sum to one.How do I compare a competitor fairly?
Compare like against like. Both rates come from the same runs, the same prompts, the same model surfaces and the same days, so a difference between two brands on one topic is a real difference. A difference between two topics is not, because the prompts differ. Watch the counts. Rates arrive as{value, n}, and n is the number of successful runs behind the number. A competitor at 100% off three runs and one at 62% off eighty are not comparable claims. A value of null means not measured — no successful run — while 0 with a non-zero n is a real, measured absence.
Changes carry {value, real}. Attensira applies a two-proportion z-test at 95% before it reports a movement. Below that noise floor you get {real: false, value: null}, which means “no change we can prove.” Do not read it as flat, and do not put it in a board deck as zero.
Finally, remember what a competitor rate is not: it is not traffic, not revenue, and not a ranking. It is the frequency with which a model, asked your buyers’ questions, brings a brand up. See Models for how the surfaces you track shape that number.