How Helply went from pre-revenue to thousands of leads and hundreds of demos before most startups have their first customer

A pre-launch AI support product with no GTM motion, no proven message, and no traction. We built five systems and ran 320+ experiments to find what worked. Here is what happened.

Company: Helply Industry: AI Customer Support SaaS Location: United States Campaign type: Full TAM outbound, LinkedIn inbound-led outbound, offer development, message-market fit testing Campaign duration: 12+ months

Results at a glance

Where Helply started

Helply is an AI-powered customer support product and sister company to GrooveHQ, a well-established help desk platform. The team came to us pre-revenue — product built, not yet in market. They had deep knowledge of the customer support space from years of running GrooveHQ, but no proven GTM motion for Helply as a standalone product.

They did not need someone to tell them about their market. They needed someone to help them find the message that would make that market move.

Before The GTM Company:

• Pre-revenue with no outbound system in place

• One previous outbound attempt — scraped Zendesk users, produced almost nothing

• Product positioned generically as a chatbot, impossible to cut through in a crowded market

• No message-market fit and no data on what would resonate with buyers

• Two clear goals: real market feedback on messaging and early demos to validate the product

"Cameron helped us set up a system that allowed us to send a massive amount of emails at scale, while still split testing ideas and messages. It has generated literally thousands of replies for us and hundreds and hundreds of demos." — Helply Founder

Our approach

The first priority before any volume went out was the offer. Sending thousands of emails with the wrong positioning is worse than sending none — it burns your TAM and poisons the data.

We helped Helply move their positioning from "we have a chatbot" to "only pay for successfully resolved tickets." A performance-based model that immediately differentiated them from every other AI support tool in the market.

Once the offer was ready, we built five systems designed to cover the entire addressable market at three tiers of personalisation — and to feed every reply and demo back into sharper messaging for the next cycle.

01 — Offer before volume No campaigns launched until the positioning was ready. Getting the message right first meant every experiment produced cleaner data.

02 — 320+ experiments, not guesswork Each message test was designed to tell us something specific. After 320+ experiments across 12 months, Helply did not have a hypothesis about what worked. They had proof.

03 — Full TAM coverage at the right tier 240,000 SaaS founders, tiered by fit and personalised accordingly. The highest-value targets got 1-to-1 manual outreach. The broad market got high-volume cold email. Nothing was wasted.

Five systems. One engine.

System A — Message-market fit testing across the full TAM

Enriched SaaS companies for their current CX tool provider. Positioned Helply as an add-on to existing tools rather than a replacement. Ran 320+ message experiments to identify what resonated with SaaS founders. Each experiment fed directly into the next — offers refined continuously over 12+ months.

System A.1 — Full TAM map, 240,000 SaaS founders

Acquired 240,000 SaaS founders from 10 data providers. Enriched each contact for tech stack and scored by fit. Tier 1 received 1-to-1 manual outreach for the highest-value targets. Tier 2 received LinkedIn and email prospecting for mid-tier accounts. Tier 3 received high-volume cold email outbound for broad TAM coverage.

System B — Inbound-led outbound via LinkedIn

Leveraged the founder's 60,000 LinkedIn followers and 200,000+ engagements. Ran all post engagements through an ICP filter to identify warm prospects. Used LinkedIn to distribute value-add assets — community offers, free tools, and frameworks. Converted organic engagement into outbound conversations at scale.

System C — Offer development

Repositioned from a generic chatbot to a performance-based model. Core offer: only pay for successfully resolved tickets. Tested offer variations systematically across all three tiers. Identified the highest-converting angle for each segment within the TAM.

The GTM Company flywheel

For Helply, the flywheel started before they had a single paying customer.

Their sister company GrooveHQ gave us real customer conversations and market intelligence. We fed that directly into 320+ message experiments, testing what would resonate with SaaS founders at scale. Each experiment told us something new. Each demo booked refined the offer further.

By month six, that compounding loop had taken them from zero traction to thousands of leads and hundreds of demos — and a product with proven message-market fit.

01 — Real world intelligence Case studies, call recordings, and customer conversations

02 — Offers and messages Positioning and copy built from what real buyers say

03 — Outbound data Reply rates, open signals, and booking data show what lands

04 — New sales conversations Warmer leads, better meetings, faster closes

Every new sales conversation feeds back into step 01. The flywheel compounds with every cycle.

The turning point

Month six was when the systems converged.

The message experiments had produced enough data to know exactly which offer, angle, and audience combination worked. From that point, scaling was a matter of volume — the playbook was already written by the data.

The results

Over the course of the engagement, Helply went from a pre-revenue product with no proven message to a company with thousands of leads in pipeline, hundreds of booked demos, and a GTM motion built to scale across their entire addressable market.

Their outbound now runs as a full TAM prospecting system — hitting every prospect with the right message based on the right signal, at the right tier of personalisation.

More importantly, they have message-market fit. 320+ experiments told them exactly what their buyers respond to. That is the kind of asset that compounds for years.

"If you are looking to get your GTM going, I cannot recommend Cameron highly enough. Him and his team are really good at what they do." — Helply Founder

What made this work

01 — The offer came before the volume Repositioning from "chatbot" to "only pay for resolved tickets" was not a messaging tweak. It was a fundamental shift in how the product was presented. That change made every subsequent experiment more likely to produce a useful signal.

02 — 320+ experiments is not a lot — it is the minimum Most companies run three or four email variations and call it testing. We ran over 320 across 12 months. That volume of data is what message-market fit actually looks like. It is not a feeling. It is a dataset.

03 — Three tiers meant no prospect was treated the wrong way A 1-to-1 manual message to a high-value enterprise prospect. A LinkedIn sequence for a mid-tier account. High-volume cold email for the long tail. Every tier got the right level of effort. Nothing was over-engineered or under-served.

04 — GrooveHQ was an unfair advantage we used properly Years of real customer conversations from a proven adjacent product gave us a head start on understanding the buyer. Most pre-revenue startups do not have that. We made sure we used it.