Inside the product

Not a dashboard with a score on it.

Most tools in this market hand you a number and a list of recommendations, then leave the work to you. This is a work surface. Every row carries the exact sentence an engine said, the source it came from, the competitor phrase that beat yours, the next move, and a button that says yes. Here is what each screen is for.

Screen one

The Answer Board

The screen your team opens. Forty buyer conversations as rows, sorted by what it would cost to win them rather than alphabetically or by score. Four states, each with one colour that means nothing else anywhere in the product.

Answer Board

Survival 2.4 of 3

14 won · 11 contested · 10 lost · 5 open

Illustrative

BUYER CONVERSATION

STATE

SURVIVAL

HOW YOU ARE DESCRIBED

NEXT MOVE

best insulated metal panel manufacturer

0.0 / 3

Not named in any engine.

Capability page plus corrections in the two roundups the engines quote.

insulated panel price per square foot

1.0 / 3

“they also sell insulated panels” · named third

Price explainer with a thickness and density table. Engines quote tables.

are kestrel panels any good

2.0 / 3

“reviews mention slow support on smaller orders”

Answer the 2024 thread publicly with current lead times.

manufacturers with their own production line

3.0 / 3

“one of the few running its own continuous line” · named first

Hold. This row survives all three turns, which is rare.

Screen two

Conversation detail, where the collapse is visible

Click any row and you see the same buyer question asked three times, across six engines, with the verbatim answer for each turn. This is the screen that explains why a row is lost, and it is the reason the board is not a dashboard. The drop happens between turn one and turn three, which is why we measure that far.

are kestrel panels any good

Turn 2 of 3

Illustrative

Turn two · doubt introduced in 4 of 6 engines

“Reviews mention slow support response on smaller orders, and lead times are not published anywhere. Vertex Insul states both openly.”

Where that came from

A trade forum thread from 2024 with no reply since, now quoted by three of the six engines. Fix the source, not the symptom.

What makes it a product and not a report

Three screens built around the conversation, not the mention.

Screen three

Approvals

Everything waiting on you, with the draft rendered as it will appear, the destination it is going to, the date it goes live and who publishes it. Three buttons: approve, edit, reject with a reason. Nothing leaves in your name until you press one.

Your rejections become rules

Screen four

The work log

Every action taken on your behalf, dated, with the measurement before it and after it, including the ones that did nothing. Three tabs: done, in progress and scheduled, so work is visible while it is happening rather than only after.

This is also your invoice

Screen five

Sources

The pages the engines actually quote in your category, which ones name you, which name a competitor, and which name nobody. Most are not your website, which is why fixing your site alone never works.

The map competitors cannot see

Screen six

Conversations

Your agreed question set, versioned and dated, with the follow up turns written out. Changing it requires written agreement from you, which is what stops a baseline being quietly moved to flatter a report.

Versioned, never silently edited

The remaining six are movement and ranges, the monthly report, signals, brand knowledge, integrations with a destinations list, and team and settings. Nothing exotic, all of it necessary, none of it billed separately.

The distinction that matters

A dashboard tells you that you are losing. This tells you why, and then does something.

What a monitoring tool gives you

A visibility score, often blended across engines
A mention count, with no sentence attached
A list of recommendations for your team to action
A monthly report you read once
The work, still entirely yours to do
No idea whether anything you did actually worked

What this gives you

The exact sentence each engine used, per turn, never averaged
The competitor phrase that beat yours, side by side
The source the engine quoted to build that answer
The next move, drafted, waiting on your approval
The work carried out by us, with your approval before anything publishes
The before and after measurement attached to every action, including the failures

Who sees what

You get the board. We get the machine room.

There are two interfaces. Yours is the screens above, written in plain words, with no mention of agents or automation anywhere in it. Ours is an operator console carrying the measurement engine, the cost model, the pattern library of what has actually worked across every client, and the margin on your account.

We tell you this because the alternative is pretending there is no machine room, and you would find out eventually. What you will never see in your product is a number of agents, because that number tells you nothing and the next company along will always claim a bigger one. What you do see is the activity trail, stamped with who did each step, machine or person.

A real Tuesday on one account is here.

See it with your own questions in it.

A twenty minute demo on your own category, or a free AI visibility snapshot on your own questions, yours to keep either way.

The product, as your team sees it

Two screens you will live in.

Everything else in the product exists to keep these two honest: the board that tells you where you stand, and the queue that will not publish anything in your name until somebody on your side says so.

The Answer Board

Approvals

And the screens that keep them honest

Conversation detail

The work log is the invoice. Agreed against shipped, with what each action moved, including nothing.

Moves

The work log is the invoice. Agreed against shipped, with what each action moved, including nothing.

Work log

The work log is the invoice. Agreed against shipped, with what each action moved, including nothing.

Shown with a sample account.