The pitch that walks in with the shopper's words
A 20-person independent brand and retail agency was invited to challenge for a 40-store grocery chain's account. The RFP scores AI, the pitch was 3 weeks out, and one billable strategist handled new business.
Pitching an AI-scored RFP with no time to find shopper language
Three narrow readers scan shopper posts in parallel, each keeping one kind of item (complaints, shoppers' own phrases, switching stories) with a link and reason, after four shared drop checks that log every discard. A fourth reader compares survivors to the client's and rivals' claims and finds white space. The strategist writes the pitch line; a rerun answers new questions.
- the drop checks threw out 22 of the 93 posts, each with its reason on file, 2 of them brand accounts written to read like shoppers
- all 70 lines on the map link back to the post they came from
- Parallel reader agents
- Rule-based post filters
- Claims sheet comparison
- Drop log
- Source-linked evidence map
- Node.js
- Command-line rerun
Same problem?Send it over.
Tell me what your version looks like and which tools it runs in. I reply with how I would build it.
-
Digital agency
The RFP response that stopped taking 6 weeks
a 6 week RFP response process became about 30 minutes per proposal
-
Advertising agency
The account defense that starts before the RFP arrives
one approved record, 9 sizes, 1.402 seconds for the still family, an average of 0.156 per size
-
Performance marketing
The retainer, on the record
all 6 approved spend changes carry a named approver, and the gate sent 1 approval back for a missing reason