“Ask Marco what the re-keying costs him a month. In his number.”
Your coach · a six-minute call
The same read in Claude and Slack at 08:52Set up once. Used all week. Coach, Do and Report, on a real week.
Replicate Labs takes your buyers, your method and your scorecard, once. From then on the coaches work the live deal, judge the email and read the week, on the platform, in Claude and Slack, and in the CRM.
Three things go in once. Then a week looks like this.
A harness is the runtime around a model: it decides what the model is fed, which tools it can call, what happens to the answer and what happens when the answer is wrong. Ours is fed your buyers, your method and your scorecard, set up with us and your methodology partner. After that nobody re-explains the business. They just ask.
In practice: your Gap Selling playbook and the discovery scorecard your managers grade on go in during implementation. Every roleplay and every deal review after that is judged against them.
Discovery roleplay against a sceptical CFO. Round two scored on your scorecard.
A six-minute call before the Northbridge meeting. The question to land, the trap coming.
Ten minutes before the renewal: the economic buyer nobody has met.
The 1:1. Bryce prepped the manager with the evidence.
The follow-up to Northbridge, judged before it sends. Paragraph two cut.
The leave-behind, built from Tuesday’s call, in the buyer’s numbers.
The proposal. Nothing priced before the buyer’s number is on the page.
#revops asks Bryce whether Northbridge is ready for commit. Not yet.
The capability map: where discovery is improving and what still stalls.
Specimen week. One rep, one deal, made up to show the shape. The blank cells are the point: the verbs fire when the work calls for them.
Twelve personas, the recorded calls, the Gap Selling playbook and the scorecard your managers already grade on.
She said the cost was “a few hours a week”. That is not a number. Go back for the number.
The same context sits behind the answer in Claude, in Slack and in the CRM. Corrections propagate; nothing is retyped.
Three surfaces. One set of context behind all of them.
The platform carries the experiences you cannot build yourself. Your own tools carry the same coach under the same seat. The CRM and whatever you build on top run on credits.
01 · The platform Where the call happens.
- A call with the coach before and after the meeting
- Roleplay against a buyer built from your personas
- Courses in your method, Dynamo in front of the room
- Bryce reading the team for managers
02 · Claude, ChatGPT and Slack Where the rep already is.
- The same coach, with the same memory, in a DM
- An email drafted in Claude and judged against the method
- A manager asking Bryce in a channel
- Nothing re-explained between surfaces
03 · The CRM, and what you build Where the harness runs as infrastructure.
- The deal read written back to the opportunity
- A call archive scored against your method
- Your own SDR agent, built on the coach’s judgement, with each draft checked against the method
- The asset built after every demo, automated
Ines, Northbridge, and what the harness made at every stop.
The grid above is the shape. This is one day of it, with the thing each surface produced. Specimen data: Ines, Marco and Northbridge are made up.
- 08:40A call, on the platform
Ines talks the Northbridge deal through with her coach before the 9:00.
Coach · call · 08:40What did Marco say the re-keying costs him? If it is your estimate, you have a meeting, not a deal.
- 09:00The buyer’s call
She asks the question the coach gave her. Marco puts a number on the re-keying.
Marco Ferrante · Ops Director · Northbridge“About four hours a day across the two sites. Call it a salary.”
Logged to the deal. Every later answer cites it. - 10:15Claude, on the seat
She drafts the follow-up in Claude. The harness checks it against the method.
To Marco Ferrante · NorthbridgeThanks for this morning. You put the re-keying at four hours a day, roughly one salary.
Our platform removes duplicate entry with one order record that syncs to every site in real time.
Is that a number your CFO would recognise, or is there one she uses?
Cut paragraph two. He has not asked what you do yet. - 14:30Roleplay, on the platform
Round two against a buyer who holds the line on price.
Coach · Round 2+240 WHeld the number and let the price wait. The buyer moved first. - 17:20Slack, the manager
Her manager asks Bryce what to cover on Thursday. The answer cites the 9:00.
#coaching · 17:20@Bryce what do I cover with Ines on Thursday?Start with the 9:00. She got a number from Marco and then used none of it in the follow-up. That gap is the 1:1.[Northbridge Sep 15 call] [Roleplay Sep 15]
On seats,
the platform.
Platform · seatsEvery rep and manager, with the coach on the deal.
- Coach calls, roleplay, courses and Dynamo
- Bryce for the managers
- The same coach in Claude, ChatGPT and Slack, under the seat
- Set up with us and your methodology partner
The cost, in his number. Everything else waits until you have it.
On credits,
what you build.
Build · creditsThe harness as infrastructure in your own systems.
- The MCP in your own Claude, with your buyers and method behind it
- A scoring job over the call archive
- The deal read written to the CRM
- Plays: the jobs the harness already runs, as a catalogue
Bring one opportunity. Watch a week happen to it.
A working session on your pipeline, in the method your team runs. You leave with the coach’s read of one real deal and a clear view of where the work would land.
