AI visibility for hunting and outdoor brands
It answers from a short list of sources, and most brands are not on it. Alces measures who gets recommended in your category, gets you into the sources that decide it, and shows you the change every month.
Free audit · No login · Four assistants, real answers
What we record
Who got named, in what order, whether the assistant actually recommended them or just mentioned them, and the pages it read to decide. Your dashboard is thousands of these.
Buyer question
“Best binoculars under $1,000 for elk hunting?”
Named, in order
Cited to decide it
Your brand is not on this list.
Why it moves
Your buyer asks
They get a paragraph with three or four brand names in it, already sorted, already justified. For a lot of buyers that paragraph is the shortlist, and the comparison shopping happens after it.
The assistant reads
A forum thread, a gear roundup, a review site, a YouTube review. Our panel records exactly which URLs each assistant pulled to build each answer, so the short list stops being a guess.
The answer follows
That is the whole leverage, and its limit. Nobody can edit the model. You can be present, credibly and on the record, in the sources it reads — and then watch what the answers do.
What changed
~9%
of search traffic already runs through AI assistants, and the share climbs every quarter. Your buyers are somewhere in that number.
Similarweb, Q1 2026
15 sites
account for roughly two thirds of everything the assistants cite. The answer your buyer gets is decided on a very short list.
5WPR citation index, 2026
3×
Brand mentions across the web predict AI visibility about three times better than backlinks. Being talked about beats being linked to.
Ahrefs, 75,000 brands
Public research, named and dated. Your own numbers come from the free audit.
How it works
We build a panel of forty real buyer questions for your product type and ask ChatGPT, Gemini, Claude and Perplexity every night. Who they name, in what order, how warmly, and which pages they read to decide.
The citations show exactly which forums, roundups, review sites and YouTube channels decide the answers in your category. That list is the work plan, and we tell you up front which of them we can reach.
We get your brand into those sources: editorial roundups, review platforms, honest and disclosed presence in the communities your buyers actually read, creator reviews, and comparison pages on your own site written the way the models quote them.
A monthly receipt: share of voice then versus now, the placements won, and the sessions and revenue your analytics attribute to AI referrals. If the number did not move, the receipt says so.
What you get
Where you stand, why, and the first placements.
Two weeks, fixed scope, one-time
Foundation included. Then the ongoing work, measured every month.
Monthly, three-month minimum
Both are quoted after the audit, once there is something specific to quote against.
What we will not do
Nobody controls what a model says, and anyone who promises a position is guessing. We guarantee the measurement, the work and the receipt.
Large studies show these do nothing, and Google says the same. We will not bill you for them.
Community presence is transparent and disclosed. Getting a brand banned from the forums that decide its category is the opposite of the job.
Optics, archery, apparel, packs, boots, stands, cameras, e-bikes and the rest of the kit. Not guns.
The Alces guarantee
Anyone selling you a position in an AI answer is guessing, and the models change under them anyway. Here is what we will put in writing instead.
Forty buyer questions, four assistants, every night, through their APIs. You see every answer, not a score we made up.
Every placement we pitch, win or lose, is on the receipt with the source it was aimed at. No line item you cannot check.
Share of voice then versus now, and the AI-referred sessions and revenue from your own analytics. If it went the wrong way, the receipt says so.
No login. Takes about ten minutes to come back.
Who you are hiring
Aiden Van Den Bush
Founder, Alces
Alces is one operator. Whoever reads your panel in the morning is the same person who writes the pitches, posts in the communities, and sends the receipt at the end of the month. There is no handoff to a junior after you sign, because there is no junior.
I built the measurement itself — the panel, the brand extraction, the source map, the dashboard you would log into. When a number looks strange I can open the raw answer the assistant gave and tell you exactly why it moved, instead of forwarding your question to a vendor.
And I hunt. I read the same forums, the same gear roundups and the same YouTube reviewers your buyers read, and I did that before it was a business. That is the part a generalist agency cannot fake: in a thread on Rokslide it takes about one post to tell who has actually been out there and who is working from a keyword list.
Or email me directly: hello@alces.app
Questions
We quote after the audit. Foundation is a fixed two-week project billed once; Program is monthly with a three-month minimum. We price against what your category actually looks like — how many sources decide it, how many of them we can reach — and the audit is what tells us that. If you want a range before you spend any time on it, email us and ask; we will tell you straight rather than make you sit through a call for it.
No, and you should walk away from anyone who says yes. The assistants change, sometimes overnight. What we guarantee is that the measurement is rigorous, the work gets done in the sources that decide your category, and you see the result on a receipt every month, whichever direction it went.
ChatGPT, Gemini, Claude and Perplexity, every night, through their APIs. We report them separately as well as together, because they do not agree with each other and the gaps are usually the useful part. We do not build the business around any one of them: the mix has already shifted once and will again.
Some of it is snake oil, which is why we list what we will not do. The part that works is old-fashioned: being mentioned in the places the models read. The 2026 studies are consistent. Brand mentions correlate with AI visibility at about 0.66, backlinks at 0.22, and around fifteen sites capture most citations. Get into the right fifteen and the answers change. Get llms.txt right and nothing happens.
Run the panel nightly and read it. Pitch the roundups and publications the assistants cite. Maintain honest, disclosed presence in the forums and subreddits your buyers use. Broker creator reviews where YouTube decides the answer. Keep your comparison and spec pages current. Then write the receipt.
Because the work is knowing which sources decide the answers in a category and being credible in them. We hunt. We read the same forums your buyers read. A generalist agency cannot post in Rokslide without it showing.
We ask the assistants a dozen real buyer questions about your product type through their APIs and record who they name and what they cite. API answers are a close proxy for the consumer apps, not a mirror, and the paid work adds manual checks in the apps themselves.
They exist, and we run small tests on our own account to understand them. We do not sell paid management yet. When we do, it will be an add-on for clients who already have the organic measurement, because that is what tells you which questions are worth paying for.
Foundation is a fixed two-week project. Program is month to month after a three-month minimum, because the placements take that long to show up in the answers. Your content, your accounts and your data stay yours.
Start here
Tell us your brand and product type. We ask the assistants a dozen real buyer questions and send you the answers, the leaders, and the sources they came from.
Run the free audit