A Series B enterprise AI platform had just raised $20M. Enterprise motion was working — good close rates, strong product — but the sales cycles were long, and the board wanted faster growth. A well-funded competitor was gaining ground, and leadership decided to go after the mid-market.
Over 20 reps were hired in about three months. SDRs, AEs, the standard playbook. Pulled thousands of contacts from ZoomInfo using basic filters — industry, headcount — loaded them into HubSpot, set up cold email infrastructure, and started dialing.
All of it came with pressure — Head of GTM was squeezed from both sides. CEO expected returns on a major headcount investment. Twenty new hires looked to him for direction — who to call, what to say, why these accounts. He didn't have good answers because the company had never actually defined a mid-market motion, not actually. Enterprise motion had years of knowledge about the buyers, but the mid-market team had a ZoomInfo export and a mandate to move fast.
So everything was either made up on the spot or borrowed from the enterprise motion. Neither really worked. Reply rates were low. Cold calls went out in no particular order to what might as well be random companies. Hundreds of dials a day into a list of thousands, with no scoring and no shared understanding of who they were actually trying to reach. Adding email personalization didn't change much — the underlying messaging didn't resonate because the team hadn't figured out who they were talking to or why these companies should care.
Meetings were booked, though. Some with the wrong accounts. Some were even closed after three months. Consistency was the issue. Some reps were delivering results, but most were focused on companies that weren't a good fit and would cost more to service than they were worth. Meanwhile, 20 new hires paid a premium (they'd overpaid to hire fast), months of ramp with nothing to show, and a growing pile of misfit deals.
But the alternative was ugly — kill outbound, move the SDRs to enterprise, try marketing-led for mid-market. Operationally survivable. But it would tell the board that the mid-market bet (their answer to the competitive threat) had failed.
We started with two decisions every rep had to make without context. Which of these accounts is worth my time today? What do I need to know before I pick up the phone?
Nobody had a clear answer. So that's the first thing we did. We ran several working sessions with their leadership and top reps to capture the judgment their best people had built up but had never written down — what makes an account worth pursuing, what signals tell you they have the problem, what changes the conversation when you reach them. Not just firmographics (they had that already) — things like what tools they were running for customer support, how their team was structured, what their public-facing support experience looked like. Each data point had a defined weight and a written rationale for why it mattered. We argued about those weights. We manually re-ran the pipeline on sample companies to check whether the output matched what a good human SDR would conclude if they spent an hour researching the same account.
Capture the judgment, define data points and weights, validate against real accounts with experienced reps.
It took about two weeks. As a result, we built an AI system that could run this research automatically (including grabbing screenshots from customers' sites) and score any company against the ICP we'd defined together — a fit score from 1 to 100, plus a short explanation of what signals were present and what they meant. Our system scored roughly 12,000 companies from their HubSpot in two days.
We then split the top segment among reps using HubSpot lists that synced directly into the outreach tool. Reps still used the same tools, same sequences. But now, before picking up the phone, they could read a document explaining why this company was on their list, what challenges it likely faced, and what mattered to them — that changed the nature of the conversation entirely.
Build the automated research & scoring system. Score 12,000 companies. Integrate into existing HubSpot workflows.
4 weeks total from kickoff to live system.
Within the first quarter, the pipeline increased 1.5x. They stopped hiring — not because of a freeze, but because the reps they already had started producing at the level leadership had originally planned to reach by adding more headcount. Monthly tool spend dropped from $20K to $10K. Half the tools they'd been paying for became redundant once the system handled what those tools were doing manually and badly.
By year-end, the mid-market team generated over $5M in revenue — more than double what the enterprise produced over the same period.
A motion that was weeks away from being shut down became the company's primary revenue engine.
As the system ran, it improved itself. It tracked which data points actually predicted fit and adjusted the weights, so the second month's targeting was sharper than the first. They started using the research documents not just for outreach but to understand their ICP more deeply — what specific characteristics made a company worth pursuing and why. When they expanded into new verticals, they didn't start from scratch.
Reps went from quantity to quality. Fewer calls, better conversations, higher output. Our system learned from what worked, and the team learned from the system.
Your conversion rates are leaving revenue on the table. Find out where — and what it's costing you.
Get in Touch →