Start with the decision.
What has to be decided, for which brand, in which category and market, over what window. Your own hunches go in too, and get tested like everything else.
A conversation, not a form
Read the market before you brief it.
Social content, comments, ads, search, audience, trends and marketplace data, read by strategists into one sourced view of your brand in the context of your category and market. Verified. Sourced. Yours.
6data layers on every capture: social, ad libraries, search demand, audience panels, trends, marketplaces
90day window with momentum and decay, so a spike is not mistaken for a trend
9instruments computed by code from the rows, every value carrying the ids it came from
0uncited lines published. A line the rows cannot support is held, not written around
Brands that move at the speed of culture grow faster. Most brands cannot see culture move.
The dashboard says what your brand posted. The agency deck says what it wants you to hear. The category, meanwhile, is being argued about in comment threads, creator videos, search bars and shopping baskets you never see. So the plan gets written from the part you can see, weeks after the moment, and the nagging feeling stays.
Athena reads all of it while it is still moving and hands you the argument, with the receipts. You walk into the room knowing what the market actually said, and able to show exactly where.
From a nagging question to a report you can defend.
Brief it the way you would brief a colleague. Athena finds where your category is really being discussed, captures it, shows you the shape of the argument before a word is written, then writes the report in front of you. You can correct it at any step, and see what every step cost.
01 · Brief
02 · Discover
Find the rooms that matter.
Athena works out where your category is actually talked about, tests every term against live posts, and drops the dead, off-market and manufactured ones. You see why each was kept or killed, and you steer.
Tested, not guessed03 · Capture
Capture what people actually said.
Posts, videos, comments, ads, search demand and shelf listings from the rooms you kept, priced before it runs and stopped at a ceiling you set. Every row lands in your category’s knowledge graph.
Priced first, then run04 · Analyse
See the argument before it is written.
Every chart, table and quote the evidence supports, laid out in the order the report will make its case, each with a note on why it is there. Add, cut or reshape before a sentence exists.
No surprises in the draft05 · Report
Get a report that survives the room.
Written part by part while you watch, numbers before the sentence about them. Every finding is tied to the posts behind it; a line the evidence cannot carry is held back. Disagree with a finding and it is re-read against the same evidence. Delivered as a page and a deck.
Defensible line by lineEvery number traces back to the post it came from.
The reason an Athena report holds up under questioning is that nothing in it is remembered or estimated. The figures are computed from what was captured, and each one carries the ids of the posts behind it. The writer works from those figures and nothing else, so when someone asks where a number came from, the answer is a click away.
01
Coverage
Every run of the capture and what it returned: rows per route, and what was not done. The methodology table writes itself.
02
Attention
The post and video leaderboard by views, likes, comments or shares, with totals per platform.
03
Voices
Verbatims. Comments and text posts ranked by likes, narrowed to a phrase when you need the ones that say it.
04
People
Who owns the posts, per platform: posts, views, likes, followers, verified, the paid share, and their top post.
05
Search
Keyword Planner volumes, related queries and people-also-ask per keyword, the top domains, and Trends as a series.
06
Ads
Ad libraries grouped by advertiser, Meta by page and TikTok by brand, with how many ads carried your keyword and one sentence on how it reads.
07
Shelf
Marketplace listings per target in position order, brand share of the top of the shelf, and the price ladder.
08
Hashtags
Tag frequency across posts and videos, with the views and likes behind each tag.
09
Timeline
Rows and views per week per platform, dated by their own time, and ad starts per advertiser per week.
Every source is tested before you ever see it.
Nothing in an Athena report is hand-picked or guessed. Athena finds where your category actually talks, tests every tag, creator and community against live posts, and links what survives into a graph you can interrogate. One real walk below: Sensodyne in Indonesia. Hover a node to see every path that reached it; click to pin. Illustrative values.
Key
Brief
Model call
Term
Account
Weak
Killed
Already in graph
Never captured
The asset you keep
Athena produces a knowledge graph of your category; every signal captured, judged and linked, vector-indexed and held in your own Postgres tenancy.
Read one before you buy one.
Athena gathers; Apollo pronounces. The capture and the instruments are hers, the verdict is his, and a report is the handoff, with his working checkable against her rows. Every report so far is on a real category and real brands, published exactly as delivered. The first three sections of each are open; the rest sits behind a work email, because the recommended moves are the part a competitor would pay for.
One question, a standing read, or the data underneath.
Every plan runs on the same capture and the same instruments. The difference is whether you want an answer, a habit, or the raw material for your own team.
Single report
$10,000per report
One question, one capture, one 13-beat report in your template. For a pitch, a launch, a board question.
- Brief to report inside the week
- Every platform discovery finds, plus ads, search and shelf
- Deck in your template plus a shareable web report
- Strategist-checked, every number cited
Programme
$7,500per report, monthly cadence
The same report on a schedule. The price per report falls with frequency, and the captures compound: month six reads against months one to five.
- Quarterly $9,000 · monthly $7,500 · fortnightly $6,500 per report
- Rolling scope, so discovery runs once
- Momentum and decay read across the whole programme
- Cancel at the end of any quarter
Data supply
$1,000per month + $5,000 setup
The capture and the lenses, without the writing. Your scope, refreshed continuously, delivered by API and MCP into your own tools, dashboards and agents.
- Setup covers discovery and scope build
- Nine instruments as endpoints, rows on request
- Reports at the programme rate whenever you want one
- Annual: $17,000 in year one, $12,000 after
Per-report rate is set by cadence: 1 at $10,000, 4 at $9,000, 12 at $7,500, 26 at $6,500. Between those points the rate steps down at the next cadence you reach.
$7,500per report
$90,000reports, year one
$0data supply, year one
$90,000total, year one
$30,000saved against buying 12 single reports
Before you ask.
Is the report written by a model?
The capture is Athena’s; the verdict is Apollo’s. The read is written from the nine instruments, part by part, with a strategist steering it. Every finding is cited to the rows behind it; a line the rows cannot support is held rather than published. Where a capture did not run, the report says so instead of estimating.
What data is in a capture?
Social from YouTube, TikTok, Reddit, Instagram, X and Twitch. Meta and TikTok ad libraries. Google search demand and Trends. Audience panels. Marketplace listings from Amazon and Steam. Not every category lives on every layer; the discovery step tells you which rooms matter for yours before we capture.
How is this different from a listening tool, or a strategy plug-in for ChatGPT?
Listening tools give you a dashboard and leave the reading to you. Plug-ins give your chatbot more data and leave the judgement to you. Athena was built and trained by strategists to give you the read itself: a sequenced argument with the moves at the end, in your deck template, with the working shown. The data supply plan exists for teams who want the raw layer as well.
Who owns the data?
The scope is yours. The captured posts stay public property of their authors and are linked, not copied. The lens values and the report are delivered to you and are not resold.
Can we start with one and move to a programme?
Yes. A single report counts as month one if you move to a programme inside 60 days, and the difference between the single and programme rate is credited.
Bring the question you have been circling.
Thirty minutes with the people who read the captures. Leave with a scope, a price, and a date for the answer.