Conversion and Engagement
Summary
The phase where acquired attention lands and is captured. Two surfaces do the work: the website and social media. The original model calls this “conversion or engagement” — worth separating the two, because they’re different jobs.
The hinge
Every acquisition channel ends the same way: the person either searches for you or follows you. This note is about what happens at that moment and just after.
Conversion ≠ Engagement
- Conversion = turning attention into a measurable action now (lead, email, sale).
- Engagement = nurturing attention over time so it converts later. They use the same two surfaces but optimize for different things. Keeping them separate stops you from judging a brand-awareness post by sales metrics, or a sales page by follower count.
Website
The conversion workhorse. Once it has enough traffic and domain authority, it becomes a testing surface:
- A/B testing — only meaningful with enough traffic; don’t test prematurely.
- Email flows — capture then nurture (the bridge into Customer database).
- Popups — capture mechanism; use sparingly.
- Powered by the inventory/price sync from The bridge into Distribution.
Instrument before you optimize
A/B tests, email attribution, and channel ROI all require one source of truth for tracking (consistent UTMs + analytics) set up first. This is an artifact-layer prerequisite — see Governing Artifacts.
Current instrumentation stack 🟢
| Surface | How it’s tracked | Status |
|---|---|---|
| Website | SOP: Stape (server-side GTM) + GA4 MCP — one-shot builds a complete website tracking mechanism | 🟢 solved |
| Social media | Zernio webhook fetches social performance data | 🟢 solved |
| Consolidation | Pulling website + social into one place — the unsolved gap | 🔴 open |
The real instrumentation problem is consolidation, not collection
Per-surface collection is handled (Stape+GA4 for web, Zernio webhook for social). The remaining problem is consolidating these signals into one source of truth — a metrics-aggregation layer along the lines of doubleloop.app. Until that exists, the analyze agent’s driver tree is reading from two disconnected places. This consolidation layer is the next instrumentation artifact to build. Tracked in Single source of truth drift.
Social media
The engagement workhorse. Its job is to maintain presence and trust: regular posting, brand updates, announcements, issue responses.
Same platform, different objective
A subtle but important point from the original model: on one platform you may run two objectives at once. Example — Instagram: Reels do acquisition (reach new people via content), while feed posts/stories do engagement (maintain the brand for existing followers). Don’t conflate their metrics.
- Link in bio is critical — it’s the literal bridge from “follow” to “convert.” Treat it as a mini-conversion page, not an afterthought.
What this phase produces
| Output | Flows to |
|---|---|
| Captured email / lead | Retention |
| First purchase | Capital (revenue), Retention (cohort) |
| Follower / subscriber | Retention |
Automation notes
- Social posting & scheduling: strong BAU automation (see Part C1 - Case Study - Peptide TikTok Network).
- A/B test execution & email flows: semi-automatable; the decision to change a page is Pilot-ish.
- Brand issue / announcement responses: keep human or human-approved — reputation risk.
Related
- Acquisition Channels (before) · Retention (after)
- Operation — feeds website inventory
- Governing Artifacts — tracking + brand context