← Back to selected work

US Coffee Tech

Signal-based outbound system for a specialty coffee tech company

Designed ICP signal architecture using Google Maps, AI site analysis, and competitor detection. Highest-value move was blocking a premature CRM integration — and coaching the RevOps lead from CRM admin into systems architect along the way.

0 → V1

outbound system designed

29k

companies in CRM audited

4 sessions

advisory engagement

Clay HubSpot Google Maps API Hunter.io Claygent AI

Context

A US specialty coffee tech company (electric roasters for artisan shops) needed to move from founder-led outbound into a repeatable system. The CRM was already in place — but stuffed with 29,000 companies, fewer than 6,000 deals, and no deduplication strategy. Leadership was about to wire Clay into HubSpot before the data hygiene problem was solved.

I was brought in for four advisory sessions to design the outbound system and stress-test the CRM integration plan.

What I built

ICP signal architecture — translated the qualitative buyer profile (small specialty coffee shops, sustainability-leaning, gas-roaster replacement candidates) into computable signals:

  • Location count — 1-4 locations is the sweet spot (large enough to invest, small enough that the founder still buys).
  • AI site analysis — Claygent reads each prospect’s website and scores for sustainability/craft messaging.
  • Competitor detection — gas roaster mentions in site copy flag a replacement opportunity.
  • Composite scoring — weighted V1 model combining the above.

Clay pipeline — Google Maps API as primary source (outperformed ZoomInfo for the SMB segment; ZoomInfo data on tiny coffee shops is sparse and stale). Three-layer enrichment with running conditions to preserve credits — cheaper signals first, expensive ones gated on prior-layer score.

The highest-value decision was a “no”

The CRM was a problem the client didn’t yet know was a problem. I audited HubSpot: 29k companies, <6k deals, no dedup logic, multiple records for the same business. Wiring Clay into that mess would have replicated noise at velocity.

I made the case to leadership: pause the integration, do controlled manual uploads first, fix dedup before automating writes. They listened. The Clay-to-HubSpot integration that would have created downstream chaos didn’t happen.

In parallel, I coached the internal RevOps lead on positioning — they’d been operating as “the HubSpot admin,” fielding ad-hoc list requests. Reframed their scope as systems architect, and gave them the audit narrative to take to leadership themselves.

Outcome

  • Client signed Clay Pro post-engagement (V1 pipeline built on Pro tier).
  • Sales team engaged with the first qualified lists.
  • The HubSpot integration was deferred until the dedup work was done — preventing weeks of cleanup later.
  • The RevOps lead came out of the engagement positioned for a strategic seat at the table, not just an operator role.

Why this engagement matters

The framing most outbound consultants miss: the highest-leverage move is sometimes blocking the work, not building it. A faster Clay-to-HubSpot pipe on broken data accelerates the problem. The four advisory sessions did three things: shipped the V1 outbound architecture, prevented a bad integration, and re-positioned the internal owner. The bill on the third item alone exceeded the engagement cost.

Let's talk

Let's build your revenue system

I'm available for GTM engineering and RevOps projects. Let's talk.