Converting cold traffic → Nautilus
Why most outbound fails,
and what a cold-traffic offer
looks like for a car wash.
A 17-minute talk from the person behind 300–400 positive replies a day, translated to how we sell Nautilus to operators who have never heard of us.
The talk in one line
“The main reason outbound campaigns don't work is you treat a warm-traffic offer like a cold-traffic offer.”
Eric Nowoslawski was Clay's 11th employee and built the content engine behind its early growth. His agency sends ~100,000 cold emails a day for clients like HubSpot, Reddit and ClickUp.
- Warm traffic needs an event on the prospect's side first — a visit, a like, a lost deal.
- Cold traffic must make someone “going about their day… become interested immediately.”
- Some businesses can never sell cold (his examples: cybersecurity, SOC 2, coaching, funeral homes).
“Using Claude to connect three APIs and launch a spam campaign doesn't make you a GTM engineer.”
His bar for the role: a specific list of plays every company should run, and the engineering to deliver value in the first touch.Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026.
First question
Can Nautilus be sold cold? Yes — but only one way we talk about it qualifies.
Sells cold ✓
“More members, more revenue from the site you already have, live in two weeks, 0% commission.” An owner can say yes to that mid-shift. That's a make-money offer.
Only sells warm
“Consolidate your tech stack.” “Replace Rinsed.” “AI-powered CRM.” These need the operator to already feel the pain. They're true, but they're not why a stranger replies.
Implication
Every cold sequence leads with the money line and a named operator result. Platform breadth, Compass, Davey and AI CMO belong in the demo and in warm follow-ups — not in the first email.
His rubric: every offer either helps people live longer, raise status, reduce risk, save time/money, or make money. In B2B cold outreach only the last one converts reliably; savings have to be “a huge amount” to work.
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026.
His cold-offer rubric
He graded every client that lasted six months against five points. Here's ours.
What converts cold
Nautilus version
verified → proposedNew Day 67% is published on nautilus.co; other customer numbers need approval before external use.
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026.
His “perfect email,” rewritten
“I saw Karen's negative review” → the Nautilus equivalent.
His (Google reputation client)
Subject: I saw Karen's negative review
Hi [name], I saw that Karen said your team arrived late and I was wondering if that's costing you customers. We help businesses remove Google reviews based on results. We just helped another roofing company in Los Angeles do the same last week. Could I call you to understand what happened?
Money (implied) · new mechanism · research · proof · one ask. He says that's the whole rubric in five sentences.
Ours (draft — Sam/Amayr to edit)
Subject: tried to buy a membership on your site
Hi [first], I tried to join [Wash] Unlimited from my phone last night and ended up at a “visit us in store” page. Operators we work with sell memberships straight from the site, tied to their [POS] — New Day converts 67% of the people who reach checkout. Live in two weeks, no commission. Want me to send you what [Wash]'s checkout would look like?
Research is real (we actually tried). The ask is a yes/no for an asset we can build in an hour.
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026.
“If you photograph one slide, photograph this one” — part 1
Warm-traffic plays every company should run. Our status on each.
His list
Nautilus
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026. Status: Monaco and Cal.com state, Oct 2026.
Part 2
Cold-traffic plays he recommends regardless of offer — and the car-wash version.
His list
- Ten minutes of real research — “the best email you can send”
- Shared school or past employer with someone on your team
- Same city: “I'm in New York, you're in New York, let's meet”
- New hires / first 90 days in role
- Tech stack named in a job description
- Hiring for a specific role
Nautilus version
- Research = try to buy. Open their site, attempt a membership, screenshot where it dies.
- Same market. “We run Surf Thru's sites in Bakersfield; you're 40 miles away.”
- New GM / new marketing hire at a chain — first-90-days message.
- Job post mentions DRB, Sonny's, Rinsed, “membership sales.” That's a stack signal and a buying signal at once.
- Shared past employer: our customers' GMs who moved to a new chain.
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026.
Engineering the first touch
He builds the deliverable before the call — for every prospect, programmatically.
His examples
- 35,000 accessibility audits of Shopify stores, each with the fix attached
- Per-company Reddit ad audits: “you have 240 active ads, these are draining budget, here's what yours would look like on Reddit”
- “GTM centers” — 3,500 verified ICP emails, 2,000 mobiles, new hires, warm overlaps, sent as the pitch
We already have the strongest version
Fast5 Xpress and Carnation Auto Spa: complete working websites with their real pricing and a live checkout drawer, built from their URL in about two hours each. Nobody else in car wash can send that.
Lighter magnets we can generate
- “Can your site sell a membership?” audit — screenshots + a conversion estimate
- Google review velocity report per site vs. the chain down the road
- Failed-payment math: members × your POS decline rate × months
His rule for choosing one: give away for free something you know other people charge for. A $5–15K website is exactly that.
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026. Build times: Nautilus Shipyard builder runs, Oct 2026.
His best advice
“Don't make the lead magnet until you know it works.”
He pitches the idea of the asset first — “I found everyone who engages with your team's posts; can I send you the file?” — and only builds once enough people say yes. “Many times we come up with lead-magnet ideas people don't want and then waste all our time making them perfect.”
Sequence this fixes for us
- Email 1: the “tried to buy a membership” note + “want to see what your checkout would look like?”
- On yes: the Shipyard build runs (≈2 hours, agent-built); reply with the live link the same day.
- On silence: a second, different cold offer — the review-velocity or failed-payment audit — never a “bumping this.”
Cost of a wasted build drops to zero; the yes-rate on email 1 becomes the metric.
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026.
Things he says not to do
Three habits that kill cold conversion.
Telling them you watched
“Don't be weird by including that bit about them being interested.” Segment on the like or the visit; write to the problem, never to the behavior.
Ours: no “saw you on our pricing page.”
Leading with savings or risk
Save-time and save-money offers need a “huge amount” to convert; reduce-risk “never sells” cold.
Ours: “cut failed payments” is a save-money line. Reframe: “recover $X of membership revenue a month.”
Spam dressed as engineering
Three APIs and a blast is not GTM engineering. The engineering is in the research, the asset, and the offer.
Ours: volume stays human-approved; the agent's job is research and assets.
Source: Eric Nowoslawski (Growth Engine X, ex-Clay #11), “Converting Cold Traffic with GTM Engineering,” Cold Outbound Clips, YouTube, Sept 2026.
Proposed · next 30 days
Four things to run, each with a reply-rate number attached.
- Rewrite the first email of the Short Question Play to the five-point rubric; A/B against the current one on reply rate.
- “Want to see your checkout?” offer as email 1 for Tier-1 accounts; build Shipyard sites only on yes.
- Closed-lost + paused re-engage sequence with a different cold offer than the one that lost.
- Competitor-post likers (Rinsed, DRB, Sonny's) via Apify → Monaco audience; message the problem, not the like.
Owners and dates to be set by Sam and Amayr. Nothing here is live yet; this is the proposal.
“Provide value to people from the very beginning as much as possible.”
We can hand an operator their new website before the first call. The rest of this deck is just making sure the email that offers it is written for someone who has never heard of us.