13+ years building growth engines, from enterprise CRM on household brands sending to audiences well above two million, to DTC programs taken from zero. I don't hand acquisition, lifecycle, and retention to three teams that don't talk. I run them as one motion, with retention as the deep end. This page answers the obvious questions so we can skip them on the call.
Sharing this up front saves us both a call. If the role lines up on the left, let's talk. If it sits mostly on the right, pass the page along to someone in your network.
Every metric below is an outcome from a program I built or ran. No composites, no extrapolations.
At MegaFood, the brand was built for women of childbearing age. The data said otherwise: the stickiest, highest-LTV customers were men 55+ buying longevity products. I moved acquisition toward who was actually buying. I read customer behavior before I read the brief, and I look for that same gap in every business I walk into.
Brand assumptions and customer behavior almost never line up. Clicks, opens, completed purchases, abandons. That's the only honest scoreboard, and I read it before the brief.
Acquisition, lifecycle, content, and ecommerce get reported separately and run by different teams. I run them as one loop — retention data tells me where the next acquisition dollar should go.
Open rates and CTR are leading indicators. Cohort retention, repeat purchase rate, and LTV are the actual KPIs. Engagement that doesn't compound to revenue isn't worth chasing.
I'm not using AI to fake competence. 13+ years of doing this work, plus the reps that come with it. AI is the layer on top.
A win-back campaign where AI surfaced 8,000 at-risk subscribers. A 0.2% click rate didn't disappoint me — it confirmed the sunset call and cleaned the list. A good disaster beats a guess.
Live frameworks that merge site sales with ESP performance into views I actually use to make calls. Product velocity, repurchase windows, inventory pacing. No SQL, no analyst between me and the answer.
When the integration layer doesn't exist yet, I build it. Custom MCP connectors, Python scanners, n8n flows. The build is the means, not the headline. I don't wait for IT to connect things.
Before a campaign ships, I have the model argue against it — "tell me why this failed" — and pressure-test the plan while it's still cheap to change. Claude, Python, n8n, MCP: how I work, not a demo I run for interviews.
Marketing leader based in Manchester, NH. I run marketing as one motion: acquisition, lifecycle, and retention orchestrated together, not handed to three teams that don't talk. I define the journey architecture, then go build it: senior enough to define the strategy, hands-on enough to execute it. I report on revenue and cohort-level performance, not vanity metrics. Client categories run from FDA-regulated supplements to intimate wellness, places where compliance and channel restrictions are operating constraints, not afterthoughts. Enterprise scale on household brands plus zero-to-scale DTC growth is a combination few CRM leaders bring to one desk.
And I don't need a roadmap handed to me. I need a leader who unblocks me when I'm stuck; the rest is mine to figure out. The thread through my career: I show up when there's nothing built yet and leave a system that runs. MegaFood (supplements, DTC, from zero to 200K+ subscribers, 150%+ YoY growth). Banter by Piercing Pagoda (SMS launch, full lifecycle rebuild). Junk Gym Digital for founder-led DTC clients since 2008. Most recently nearly five years as CRM Engagement Manager at Signet Jewelers (through May 2026), directing retention and email for Kay Jewelers, Jared, and Peoples Jewellers across the US and Canada, with campaigns reaching audiences well above two million. Now looking for the next room with nothing built yet.
Stack: Klaviyo · Omnisend · Attentive · OrderGroove · Yotpo · GA4 · Adobe CDP · Claude · Python · n8n · MCP
Junk Gym Digital is my consulting practice. Three to five clients at a time, retainer or project. Same operator, same skill set, narrower scope. If a full-time hire isn't on the table right now (timing, budget, freeze), this is the lower-friction first step. The full-time door stays open either way.
junkgym.com/work-with-meTell me about the role, or forward the link to the hiring manager. If it's a fit, you'll know on the call.