M MemberIntel KB

reference

Elevator Pitch — What MemberIntel Is

Plain-language descriptions of MemberIntel for different audiences — a one-liner, an elevator paragraph, and riffs tuned to designers, engineers, friendly outsiders, and curious customers. Use when explaining the project to someone new in under a minute.

The canonical pitch — read this when you need to explain MemberIntel to someone in under a minute. Each version is plain-language, no jargon. Pick the one that fits your audience.


One sentence

MemberIntel is an AI advisor for membership-site operators — a smart sidekick that watches their site and tells them what to charge, who’s about to cancel, and which content is actually working.

One paragraph (the canonical elevator pitch)

MemberIntel is an AI advisor Blair’s building on top of MemberPress. Imagine every membership-site owner has a smart sidekick that watches their site and tells them what to charge, who’s about to cancel, and which content is actually working. Each customer gets a private AI that learns their site, plus a shared playbook built from anonymized patterns across everyone else. Free tier runs on a local model to keep costs down; Pro is $29/mo on Claude. We’re a small team scoping in May, building in June, launching in October. The trick is making it feel like a real SaaS product — not a plugin admin screen — and keeping it brand-flexible because V2 expands beyond MemberPress to platforms like BuddyBoss.


Audience-tuned riffs

For a designer

Same product. You’d own how the whole thing looks and feels. The hard part is that it has to feel like a real SaaS product — not a plugin admin screen — and stay brand-flexible because V2 expands beyond MemberPress. First job is turning a rough Claude mockup into a real Figma file. Brain editor and onboarding wizard are the meaty UX problems after that.

For an engineer

Same product. RAG on Anthropic Sonnet (Pro tier) or a local Ollama-class model (Free tier). Per-tenant brain isolated via Row Level Security on Postgres; global brain built through a cross-pollination pipeline with k-anonymity floors and explicit permanent-exclusion for sensitive contexts (medical, advocacy). GCP hosting, pgvector for retrieval. Python 3.12 + FastAPI behind a thin llm.call wrapper; all model routing is server-side, never user-configurable. Cost discipline is a first-class constraint, not an afterthought.

For a friendly outsider

Same product. You know how every membership-site owner spends Saturday morning poking around their admin panel trying to figure out which subscribers are about to cancel, which course nobody finishes, whether they should raise prices? MemberIntel is the thing that just tells them. Built on MemberPress for v1; expanding to other membership platforms after that. Free tier for everyone, $29/mo if you want the smart model.

For a curious customer

Imagine if every time you logged into your MemberPress admin, there was a smart advisor in the corner that already knew your site, already saw the patterns in your members’ behavior, and could just answer questions: “Should I raise my prices?” “Who’s about to cancel?” “Which content is actually working?” That’s MemberIntel. Free tier costs nothing; Pro is $29/mo for the smarter model and live data. Your data stays yours — only anonymized patterns are ever shared, and sensitive site categories are excluded by design.


Usage notes

  • In conversation: use the one-sentence version. Save the paragraph for follow-up questions.
  • In a Slack DM or email: paste the audience-tuned riff that fits.
  • In a deck or marketing draft: start with the paragraph; let the deck do the rest.
  • Don’t paraphrase too much. The phrases “smart sidekick,” “watches their site,” “what to charge, who’s about to cancel, which content is actually working” have been tested in conversation — they land. The longer technical phrases (RAG, k-anonymity, cross-pollination) only belong in technical audience riffs.

The pitch evolves. When language stops landing in real conversations, update this page rather than letting the doc drift away from how the team actually talks about the product.

For: S Seth Shoultes B Blair Williams S Santiago Perez Asis A AI Engineer R Russell P Product Lead