#1355 1988 · Custom Research Inc. · Professional services
A research firm cut its client list from 138 to 67 — and doubled revenue with the same staff
the problem
A hundred-person firm spread across 138 clients was busy everywhere and deep nowhere, with growth stuck at the staff's capacity ceiling
background
Custom Research Inc., a Minneapolis marketing-research firm run by partners Judith Corson and Jeffrey Pope, served 138 clients in 1988 with roughly a hundred people. In a professional-services firm that shape is a treadmill: every small client carries setup cost — learning their business, courting, proposals — that never amortises, and the firm's best people are scattered across one-off projects instead of compounding knowledge inside a few relationships. Growth means more clients, more scatter, more hires; margin never improves.
The industry-standard answer was more selling. CRI's steering committee concluded the opposite: their scarce asset was not leads but attention, and the deliberately unpopular move was to stop serving most of the people willing to pay them.
what everyone would do
Grow by addition: hire more researchers, chase more accounts, take every project that covers marginal cost. Headcount and revenue rise in lockstep, margin doesn't move, and the firm's knowledge stays a mile wide and an inch deep — the standard professional-services treadmill.
what they saw
138 clients wasn't a customer base — it was 138 overheads. Serving half as many let the same hundred people go deep enough to matter, and depth, not breadth, is what doubled revenue without a single hire.
the move
From 1988 CRI concentrated on high-volume, repeat clients and let the rest go — the roster fell from 138 clients to 67 by 1995 while the count of large accounts rose from 25 to 34. The freed capacity went into depth on the survivors: teams organised around each client, requirements distilled to four measurable essentials (accurate, on time, on budget, meets expectations), end-of-project satisfaction surveys on every job, and an unconditional guarantee. The arithmetic did the rest — revenue doubled with no increase in staff, and the deepened relationships won the firm the 1996 Baldrige Award as documented in NIST's profile.
why it works
Relationship costs are step-fixed per client, so halving the roster returns a huge block of capacity at zero cash cost — and reinvesting it in the surviving accounts compounds: deeper knowledge wins bigger projects from the same buyers, which raises revenue per relationship, which justifies still more depth. Concentration also makes quality measurable (four essentials per project, surveyed every time) in a way scatter never permits, and the clients who remain feel the difference — they're getting the attention 70 others used to consume.
the payoff
Clients 138 → 67, large accounts 25 → 34, revenues doubled with unchanged headcount (NIST, 1996 Baldrige profile); $21M revenue placed CRI in the industry's top 40.
where it breaks
It fails when concentration curdles into dependence — fewer clients means each departure is a crater, so the kept accounts must be genuinely diversified and growing. It fails if the freed capacity leaks into slack instead of depth (the cut must come with a system for reinvesting attention), and in businesses where small clients are tomorrow's large ones, culling by current size amputates the pipeline.
what came after
Became a reference case for the focus strategy in professional services — the documented instance of 'fire your small clients' actually run to completion and audited by a national quality program.
references
- [1]Baldrige Award Recipient profile: Custom Research Inc. (1996)NIST / Baldrige Program, 1996nist.gov
- [2]1996 Baldrige Award WinnersQuality Digest, 1997qualitydigest.com
- [3]Choose or Lose — how CRI grew by cutting its customer base in halfInc., 1998inc.com