Connect Reckon to Claude, ChatGPT, Cursor or whatever you already work in. It reads your Meta and Google performance, what each ad actually says, and Reckon’s call on every one of them, judged against the other ads selling the same product. Reckon’s finding arrives already worded, so your assistant relays a conclusion instead of inventing one from a table.
“Next month's brief is attached. Reckon says to replace our best seller's ad rather than cut it. Why, and what should the replacement keep?”
In the assistant where the brief already lives. No export, no screenshot.
Replace, not cut. It costs 3.1 times what the other ads selling the same product cost, but it is the only ad still reaching that audience.
30 days of settled data, against the 11 other ads selling the same product. Both ads beating it open on a wearer. This one opens on the product pan it launched with.
Keep the offer and the audience from the brief, reshoot the first three seconds on a wearer, and run it beside the original for a week.
Reckon reads your creative and judges every ad before you ask, and stores the result. The server hands that over: the numbers, the evidence and the call, already worded. No second AI sits between your assistant and your data, so your assistant reasons over it and writes the reply, and anything it adds beyond Reckon's finding is its own suggestion.
When you want to look at the evidence yourself.
Every table, creative and verdict, with nothing between you and the numbers.
When you want every figure in the answer checked.
It lives inside Reckon, and nothing reaches you until each number in it has been matched to the data it came from.
When the answer has to meet work that lives somewhere else.
The brief, the shoot budget and the client deck are already in your assistant. Reckon serves the findings straight from what it has stored, and your assistant takes it from there.
Meta ships an MCP server, it is free, and it is the obvious thing to connect first. It gives your assistant direct access to the ad account: the numbers, and the controls. Reckon gives it something different, which is what Reckon has already worked out from those numbers and from the creative itself.
Compared against the tools Meta's own MCP server published as of September 2026.
The same questions the dashboard answers, in the same shape, with the same gates. Each answer names the accounts it read, the window it covers and anything it had to leave out.
The call
Reckon's verdicts on your ads, with the evidence each was judged on.
Your highest-stakes decisions first
Ads ranked by the money at stake, each told to scale, hold, protect, cap, trim, sunset, replace or cut, with the evidence behind the call and any gate that held it back. One answer carries the top of that queue and says how many more there were.
What changed since last week
Yesterday against the day before, the week's direction, where spend concentrated, what launched and what is going tired.
Creative patterns
What kind of ad works for you, not just which one did.
Which hooks, angles and formats win
Every creative attribute ranked by the metric you care about, with a gate on any row too thin to call.
The ads behind a winning pattern
The individual creatives carrying one answer, for gathering references before a brief.
Find an ad by what it says
Spoken lines, headlines and calls to action, searched across the tagged library. The answer says when only part of it was read, so silence is not mistaken for proof you never said it.
Performance
Meta and Google in one shape, on settled days.
What you spent and what it returned
The headline totals for a window, per account and combined, naming every figure it could not honestly fold together.
Ranked ads, ad sets and campaigns
One table sorted across every account, on the metrics you asked for, rather than a page per account.
What every conversion event is worth
Subscriptions, trials, registrations and custom app events with their count, their value and the spend behind them, which the headline revenue and return figures count only purchases for.
One metric over time
Dated points by day or week, so a trend is read rather than guessed at.
One ad, in full
What the creative says and shows: its tags, the spoken hook, the call to action and the transcript where one exists.
Where an ad wins and loses
Hook, hold, click, convert and engagement as percentiles against the other ads on the account, so a weak stage is named.
Competitors
The rivals you track, read the way Reckon reads you.
Who you are watching
The competitors this workspace tracks, and whether Reckon has resolved each one.
A competitor's live ads
Their running creative, the copy and the formats, with how long each has been up. Nobody publishes spend, so run length is the only signal and it is a weak one.
Angles they run that yours do not show
One creative question asked of both libraries at once, in both directions. Their side is everything Reckon has tagged and yours is the window you choose, and the answer states both rather than reading as like for like.
Audience
Where delivery actually went.
Delivery, placements and targeting
Broken down by age, gender, placement, geography or device, with the saved audiences an account holds and how one ad set's targeting has moved.
The workspace
What is connected, how current it is, and how much Reckon has read.
Every connected account
Each account with its channel, currency and newest day of data, the products the workspace sells, and the competitors it tracks.
Whether the data is still landing
Which accounts are mid-sync and how the recent ones ended, so a thin answer can be told from a quiet week.
How much creative Reckon has read
Coverage of the creative library, so an empty pattern is not mistaken for an unused one.
Who changed what, and when
Budgets raised, ad sets paused, targeting edited. When a number moves on a single day this is usually the reason, and nothing else can see it.
Brand and boards
The positioning and the reports your team keeps.
Positioning, personas and voice
The brand as the team wrote it down, so a suggestion comes back in your own language.
The reports your team keeps
Saved boards, which are a fair map of what this team watches week to week.
One board in full
Every report on it, with the accounts, window and metrics it was scoped to.
Not “summarise my account”. The ones with a shoot, a budget or a client call riding on them. Copy any of these into your assistant once it is connected.
Your assistant can recommend a change. It cannot make one: nothing here can pause a campaign, move a budget or publish an ad, so what to act on stays your call.
Which ads should I turn off this week, and where does that money go instead?
Reads The call
Which product are we spending real money on where every ad selling it is weak?
Reads The call, Performance
Which ad is strong enough to build a whole batch around, and what should the next five versions keep and change?
Reads The call, Performance, Creative patterns
Our best ad stopped working. What in the data is consistent with people tiring of it, what else could explain the drop, and what can the data not tell us?
Reads The call, Performance
Which ads lose people in the first three seconds even though the ones who stay end up buying?
Reads Performance
Rank the angles Reckon has tagged in our last 60 days by cost per sale, say which rows are too thin to call, and name an angle that does not appear in that window at all.
Reads Creative patterns
Which messaging angles do the brands we track run that do not show up in our last 60 days? Give them to me as test candidates, and flag any factual claim a human has to substantiate first.
Reads Competitors
Write Monday's brief: what changed last week, what we should shoot this week, and the numbers that argue for it.
Reads The call, Creative patterns, Brand and boards
You need a Reckon workspace with at least one ad account connected. Get early access if you are not there yet.
Paste it into your assistant's connector settings, or run one command.
Your assistant opens Reckon's sign in. No key to copy, no secret to store.
Choose which workspace to share. That choice is what the connection can read.
Connecting is not the same as having data. Check which accounts it can see, how current they are, and how much creative Reckon has read.
Works on every Claude plan, including Free with one connector. On Team and Enterprise an owner adds it first under Organization settings, then Connectors, and each member connects from Customize.
Claude's guide to custom connectorsWhich ad accounts can you read through Reckon, how current is each one, and how much of our creative has Reckon analysed so far?
A good answer names your accounts, the newest day Reckon holds for each, and how much of your creative it has read. If it does, you are ready for the questions above.
No tool can pause a campaign, move a budget, publish an ad or change anything in your workspace. The server has no write path at all, and runs no AI of its own: it answers from what Reckon has already stored, and never asks Meta or Google for anything on your behalf.
You choose the workspace when you sign in, and the connection can read all of it through these tools: performance, creative text, brand notes and boards. Naming an account in a question narrows that question, not the grant.
Reckon checks membership on every request, so someone removed from the workspace loses access within about a minute. That is the fast way to cut someone off. Disconnecting the assistant stops new sign ins, but a token it already holds can keep working for up to a day.
Reckon records which assistant read what, and when. There is no screen for that yet, so ask and we will send it. Limits on how much one workspace can ask at once protect the dashboard, which shares its servers with these reads.
Answers travel to whichever assistant you connected, and what it does with them is governed by that product's own terms, not by Reckon's. Ending access stops future reads. It does not recall what was already delivered.