B2B SaaS United States

Past the referral plateau with five GTM systems

New deals stopped depending on referrals alone. Search pages, the newsletter and outbound each became a channel the team could measure, budget for and plan around.

Illustrative scenario — directional numbers · not a named client. An early-stage B2B AI company, described by category rather than named.

Illustration — not a product capture

Azivra 01 The challenge

The challenge

The company sold AI-driven reporting to growth teams at B2B software companies, and almost every customer had come through the founders’ network, referrals or word of mouth. That carried it to somewhere between $40,000 and $60,000 a month, and then growth went flat. Referrals couldn’t be forecast and weren’t bringing in enough new opportunities, but every new channel looked uncertain, so the team kept leaning on the ones it already knew.

The strategy

Sales Funnel Management
AEO — SEO for LLMs
Lead Generation

We started at the bottom of the funnel: a sharper offer, a sales process the team could repeat, and a record of why each prospect said yes or no. New sending domains started warming up on day one, because that can’t be rushed. Then the new channels went in, each run as a measured experiment: one founder publishing on LinkedIn, a weekly newsletter for a single buyer persona, comparison, alternative and use-case pages written for search and answer engines, and outbound on five sending domains with 15 warmed-up inboxes. Every channel got a fixed budget and its own metrics before it got more money.

The playbook: 5 GTM systems every early-stage B2B company should build

How it ran

Weeks 1–3: fix the close first

Before adding a single channel, we went through every deal from the previous two quarters with the founders, won and lost.

The offer was rewritten as one page: who it’s for, the problem, the outcome, the price and the timeline. The two founders had been describing it differently on calls.
Discovery calls got a fixed outline, and the pipeline got five stages with exit rules, so “qualified” meant the same thing every week.
From then on, every deal got a one-line win/loss note.

The notes changed the target. Most of the lost deals were “not now” from companies too small to have the reporting problem yet, so the ideal customer profile moved up a size band before any outbound was written.

Weeks 1–6: outbound infrastructure, warming in the background

Warm-up can’t be rushed, so it started on day one, alongside the sales work:

five sending domains, kept apart from the main company domain
15 inboxes, authenticated with SPF, DKIM and DMARC and warmed up gradually
two LinkedIn profiles: the founder who sells, and the head of sales

Weeks 4–12: distribution and search

Founder-led LinkedIn: one founder posting two or three times a week about the reporting problems growth teams actually run into, and spending time every day in the comments of the people the company wanted to sell to.
A weekly newsletter for heads of growth: one automation workflow, one attribution breakdown and one question to reply to.
High-intent pages: comparisons with the two tools prospects mentioned most on calls, an alternatives page, and use-case pages built from the problems in the discovery notes. Each page opened with a direct answer, so answer engines had something to quote.

Weeks 6–12: outbound, one segment at a time

Outbound started with one segment and one message: growth teams that had just posted a marketing operations role, often a sign that reporting had broken. Emails were short, opened with an observation about the company and ended with a small question. Volume went up only once that segment and that message were getting positive replies.

How each channel was judged

Every channel had a written experiment before it started: a hypothesis, a fixed budget, a time box and a decision rule. The team watched early numbers on purpose:

first conversations by source, with “How did you hear about us?” asked on every call
positive replies and meetings, by outbound segment
replies to the newsletter, and sales calls that mentioned it
demo requests from the comparison and use-case pages

By the end of the quarter, outbound and the high-intent pages were producing conversations the team could plan around, so both got more budget. Partnerships needed longer than a quarter to show anything, and stayed a small bet.

What another team can take from it

Fix the close before you add channels. Every lead bought before then costs more than it looks.
Start the slow parts on day one. Domain warm-up and an audience both take weeks.
Judge each channel by the earliest number that predicts revenue, and agree what “working” means before you start.

The full method is in the playbook this scenario applies.

Azivra 02 The results

The results