Who Makes the Cut: A Two-Filter System for Selecting Certification Partners

Twenty-two strong applications. Three maybes. A cohort target of 25. Sooner or later, every founder building a certification network stares at this exact spreadsheet — and feels the same gravitational pull toward rounding up.

Resist it. The empty seats will do more for your brand than the marginal partners ever could.

Knowing that in the abstract is easy. Acting on it under cohort pressure is not — unless you have a selection system that makes the decision for you. The strongest one I know combines two frameworks that were never designed to work together: David Baker's market-viability tests, which measure whether a candidate's practice can succeed, and Michael Port's Red Velvet Rope Policy, which measures whether the person behind the practice belongs in your community.

Baker studied 1,340 expertise firms and found one trait the durable ones shared: they were ruthlessly selective about who they associated with. This article turns that selectivity into a repeatable two-filter process you can run on every applicant.

When One Filter Isn't Enough

The Candidate Who Passed Every Test Except the One That Mattered

I once saw a founder admit a partner whose application was flawless by every analytical measure: a well-defined niche, real depth of expertise, a large addressable territory, healthy revenue. Half a year on, that same partner had picked fights with three fellow practitioners, trashed the methodology in public LinkedIn posts, and was demanding bespoke changes to the diagnostic tool — changes that would have broken it for the rest of the network.

The application process hadn't failed. It had only ever been built to answer one question: can this person's practice generate results with our methodology? Nobody had asked the question Port puts before everything else — will this person leave the ecosystem better than they found it?

Run only the analytical filter and you'll admit qualified people who poison the well. Run only the human filter and you'll admit lovely people whose practices structurally cannot win. A serious selection process checks credentials and character — skipping either is how partner networks end up with a toxic star or a friendly failure.

So: two filters, run in sequence, both mandatory.

Filter One: Can They Win?

Baker's Market-Viability Tests, Applied to Applicants

Baker built his positioning tests to assess whether an expertise firm has the structural conditions to thrive. Point those same tests at a certification applicant and they tell you whether the candidate's practice has a real market underneath it — independent of how enthusiastic the candidate is.

The insight inventory. Ask the applicant, live and unprepared, to give you 20 observations they could only have earned by working repeatedly inside their niche. Not recycled business-book wisdom — earned pattern recognition. Someone with genuine depth keeps going without effort: how manufacturers with 50-200 employees keep fumbling the engineering-to-production handoff, how a cancelled operations meeting three weeks running is the earliest warning light, how seasonal hiring drives a Q1 spike in quality problems. Someone whose expertise is borrowed from reading rather than lived runs dry around insight number 7. Baker treats this pattern recognition as the bedrock of real expertise, and the test exposes its absence in minutes.

The competitor census. Have the candidate state their niche, then count who else serves it. Under 10 competitors and the market is too thin to feed a practice. Over 200 and the candidate hasn't actually specialized — they're fighting everyone and visible to no one. The viable band sits between 10 and 200: enough buyers to matter, few enough rivals that depth genuinely differentiates. "I help businesses grow" puts a candidate up against thousands. "I help mid-market healthcare organizations redesign patient intake operations" might face 40 — a fight that candidate can win.

The list test. Could the candidate purchase a mailing list of prospects in their niche? It sounds blunt, and Baker considers it one of his most dependable signals. A list broker carrying that category means someone has already mapped and segmented the buyers — the market is defined. No list, and systematic business development has nothing to aim at.

The specialist test. If the candidate needed to hire a dedicated specialist in their niche, could they find one? This proxies for market maturity. Where specialists exist, there's a recognized body of knowledge, a career path, and steady demand. Where none exist, the niche may be too new — or too narrow within a narrow field — to carry a partner practice.

The geography test. Is the candidate's territory big enough to sustain a full practice over years, not months? A restaurant consultant confined to one small city exhausts the local client pool within a year; the same consultant working a metro of 5 million has decades of runway. Remote delivery has stretched this constraint enormously, but it hasn't erased it — some work still demands proximity, and even fully remote practices need enough market density for referrals and repeat business to compound.

Pass these five and you know the candidate's practice rests on solid commercial ground. You still know nothing about what they'll be like to share an ecosystem with.

Filter Two: Do They Belong?

Port's Red Velvet Rope, Adapted for Partner Selection

Michael Port built the Red Velvet Rope Policy for a problem analytics can't see: people who are professionally qualified and personally corrosive. Right niche, right depth, right market — and a trail of drained rooms behind them. Port scores the human dimension across five criteria; here they are, adapted for evaluating partners:

Vertical depth. Does the candidate bring existing relationships in a specific segment? Fifteen years of trust built in financial services converts to revenue far faster than a vague intention to "explore" financial services. Established relationships mean referral sources and a reputation your certification can amplify — instead of having to manufacture from zero.

Drive. Will they go out and build a practice, or collect the certificate and wait for the phone to ring? A network cannot subsidize passive members. Every seat you fill with a passenger is a seat you denied to a builder — that opportunity cost is real even though it never shows up on an invoice.

Alignment. Do they actually believe in what you're building, or are they shopping for a LinkedIn badge? Mission-aligned partners hold on through the hard stretches — thin pipelines, methodology revisions, growing community obligations. Badge collectors defect the moment something shinier appears.

Coachability. Will they deliver the methodology as designed, at least until they've earned the standing to customize within guardrails? Watch the application process itself: a candidate arguing with core elements of the framework before they're even admitted will argue ten times harder once the certificate is on their wall.

Energy. After an hour with this person, do you feel charged or depleted? Yes, it's subjective — intentionally. Your gut response tends to predict the community's response with uncomfortable accuracy. A partner who drains you will drain the other partners too, and that kind of erosion is slow, invisible, and ruinous.

"Capability gets a candidate to the door. Character is what gets them past the rope. Admit no one until both answers are yes."

Make the scoring mechanical so cohort pressure can't bend it: rate each of Port's dimensions from 1 to 5, and require 75% or better on both filters. Concretely, that means at least 15 out of 25 on the human criteria and passing no fewer than 4 of Baker's 5 market tests. Below either line, the answer is no — however much you want the headcount.

Selecting a Portfolio, Not Just People

Balancing Vertical and Horizontal Positioning Across the Cohort

There's a third layer of evaluation most founders miss: how each candidate fits the cohort as a whole. Baker draws a line between two positioning strategies your applicants will arrive with. Vertical positioning is industry-bound — healthcare, manufacturing, financial services. His data puts 85% of successful expertise firms in this camp, and the logic is obvious: you know precisely which conferences, publications, and associations hold your buyers, and every engagement deepens industry knowledge that compounds. Horizontal positioning is discipline-bound across industries — data governance, leadership development, operational excellence — trading some focus for variety and resilience when any single industry turns down.

A founding cohort needs a deliberate blend of both, for three reasons:

• Referrals without rivalry. A healthcare vertical and an operational-excellence horizontal run complementary, non-overlapping practices — they can pass each other work without territorial flinching. Stack 25 partners into the same industry and you've built internal competition; mix 5 healthcare verticals with 5 operational horizontals and you've built collaboration.
• Richer intelligence. Verticals feed you deep industry signal; horizontals spot patterns that cut across sectors. Together they produce benchmarking data and methodology refinements neither group could surface alone.
• Spread risk. An all-vertical cohort concentrated in one industry contracts whenever that industry does. A blended cohort spreads economic exposure across several markets.

So map every applicant against the cohort you already have. Three financial-services partners on board? The fourth financial-services applicant may be worth less to the ecosystem than a somewhat weaker candidate from an industry you don't yet cover. You're not just picking individuals — you're composing a portfolio.

The Strategic Value of No

Why Empty Seats Beat Marginal Partners

Back to that spreadsheet: 22 strong, 3 maybes, target of 25. Here's what admitting the maybes actually costs.

First, it devalues every yes you've already given. Each marginal admission tells the partners who earned their place that the bar they cleared and the bar the maybes cleared are the same bar. Baker's research on long-lived expertise firms points one way on this: the ones that endured guarded their associations relentlessly. Your best partners will register the dilution — quietly at first, then with their feet.

Second, the damage compounds. In Modern Monopolies, Alex Moazed shows that networks are path-dependent: the earliest participants fix the quality standard for everything that follows. A founding cohort with mediocre members stamps that signal onto every future cohort's reputation. Starting strict and loosening later is straightforward; starting loose and tightening later is close to impossible.

Third — and this is the part founders underestimate — rejection is marketing. When you decline qualified applicants, the market hears about it. "They turned down 100 applicants to accept 22" positions your program more powerfully than anything your copywriter could draft. And the people you declined don't vanish; they come back as Cohort 2's most prepared, most motivated applicants.

Launch with 22 excellent partners rather than 25 adequate ones. Those three vacant seats aren't a shortfall — they're evidence that admission means something.

None of this is bureaucracy. The two-filter system is brand protection: every partner you admit carries your methodology into the market, and every partner you turn away protects the ones who made it through. Build the filters before the first application arrives, apply them without exception, and trust that the candidates worth having will clear both.