Product / Competitive intelligence

Find out which companies get picked instead, and what they are cited for.

Knowing a competitor appears more often is not useful on its own. What you can act on is the list of questions you lose, the sources the winning answers cited, and the claims those companies are associated with that you are not.

Check my AI visibility See a sample report

Quick answer

Competitive intelligence compares you with the companies that appear in the same AI answers. It reports share of voice by intent, the buyer questions you lose, the sources behind those wins, the claims engines associate with them and not you, and competitors found inside your own answers.

What it shows you

All of it comes from stored answers, so every count can be opened and read as the text an engine actually produced.

  • Share of voice, per question set

    How often you are named and recommended next to the companies that appear in the same answers, for the questions that were actually asked. Grouped by intent, so a comparison question and a best-of question are not averaged into one number.

  • The questions you lose

    The specific questions where you are absent or ranked below someone else, with the answer text and the companies that took the slot. A list of questions is something a writer can act on; an aggregate is not.

  • The sources behind their wins

    The pages the winning answers cited. This is usually the most useful part: it tells you whether the win came from their own site, a review site, a listicle or a forum thread, which decides what you do about it.

  • Claims you are missing

    The claims engines associate with a competitor and not with you. Some are things you do not do. Some are things you do and have never said in a place an engine can read, which is the fastest brief you will get.

  • Competitors you did not name

    Companies that keep appearing in your answers without being on your list. You confirm or drop them, so the comparison set is the real one rather than the one from your last board deck.

How the comparison is built

Three things have to be true before a competitive number is worth reading.

  1. You confirm the comparison set

    Onboarding proposes competitors from your site and from who actually appears in your answers. You edit the list. Anything discovered later is flagged rather than added silently.

  2. The same questions run for everyone

    Comparisons only mean something when the question, the engine and the date are held steady, so the counts come from the same stored answers you can open and read.

  3. The reasons come from the evidence

    Counts, positions and cited sources are recorded, not estimated. The written explanation is generated from those records, and each part of it points back at the answers it came from, so you can disagree with it.

Engines are not limited by plan. Every plan asks every engine you have connected, so a cheaper plan never means a narrower view of the market.

How to read it honestly

Three limits that change what you should conclude.

  • A competitor appearing more often is a result, not a cause. The cause is usually in the sources column, which is why the sources are shown next to every win.
  • Absence across the questions we asked is evidence about those questions. It is not proof that no engine ever names you.
  • Engines personalise and change. Two runs of the same question can disagree, so a trend across repeated runs says more than any single comparison.

Planned, not available yet

These are on the roadmap and are not part of the product today. They are listed so nothing above reads as more than it is.

  • Modelling where in a purchase a question sits, with personas walking it, is planned. Today questions are grouped by intent, which is the evidence that modelling would run on.
  • Ranking actions from your own past results is planned. Scoring today weighs the evidence behind an action, and is not trained on your history.

What these results are, and are not

Everything the product reports is a snapshot of the questions and engines that were tested, on the day they ran. AI answers change between runs, accounts and regions, so results vary and nothing here is a ranking. We do not guarantee that any engine will mention, recommend, cite or include you.

ModernTechLap measures and influences the evidence around a business; it does not control AI model outputs.

Common questions

How do you choose which competitors to track?
Onboarding proposes them from your own site and from the companies that actually appear in your AI answers, and you confirm or edit the list. Competitors that start showing up later are flagged for you to accept rather than added on their own.
Is the explanation of why we lose written by an AI?
The evidence is not. Mentions, recommendations, positions and cited sources are recorded from the stored answers. The written explanation is generated from those records and links back to them, so every sentence in it can be checked against an answer you can read.
Does a cheaper plan track fewer AI systems?
No. Engines are not limited by plan. Every plan asks every engine you have connected, so the competitive picture is not narrower on a smaller plan.
Can you tell us why a model prefers a competitor?
We can show what the answers said, which companies were named, and which sources those answers cited. We cannot show the model's ranking logic, because nobody outside the model vendors has it. Anything beyond the recorded evidence would be a guess.

Find out what AI says about you. Then change it.

Free baseline, 25 buyer questions, every AI system you connect. About ten minutes, no card.

Run your free baseline

25 buyer questions across the AI systems you connect, for .

Checking …