The Partner Overlap Playbook: Best Customers to a Ranked Recruit List

This playbook is for founders, account executives, SDR leads, and partnerships teams who need to find the agencies, consultancies, and implementers that already serve their best customers before any of those firms has agreed to share data. It takes a best-customer list and produces a ranked list of 20 to 40 service firms. Each firm gets a counted public-evidence overlap score, a persona check, a named practice lead, and a place in a recruit cohort sized from published activation rates. Plan on about two working days. That target is mine: no published source times the individual steps.

What an ideal partner profile must contain

An ideal partner profile is a short written list of the traits a service firm must show before you recruit it. You must be able to observe every trait on it from outside the firm, before anyone signs. A trait you can only learn after signing belongs in onboarding, not in the profile you source against.

Forecastable builds the profile around four dimensions. Here is what each one looks like on the public web:

• Account overlap with your target market. Client logos and case studies that name your best-fit customers.
• Segment and vertical fit. The industries and company sizes that keep showing up in those case studies.
• Motion fit. Whether the firm would resell, refer, co-sell, or build on your product. A listing in an adjacent vendor's reseller directory points to resale. A services-only site points to refer or build.
• Selling capacity. Certified-staff counts and open job posts for the service line that would carry your product.

Add two traits the four dimensions miss:

• Presence at the buying committee. Forecastable names this as a trait that predicts partner production. The firm must work with the function that signs your deals, not just with the same companies.
• No competing service line. One partner-profile guide warns that a firm whose services overlap yours is "not going to refer business they could keep themselves."

The profile matters because broad recruiting is expensive and most of what you recruit will not produce.

Partner overlap without account mapping: why public evidence comes first

Account-mapping tools measure overlap by matching two companies' customer records, so they need a partner who has already agreed to share data. No prospective partner has agreed to that before the first call. Estimate overlap from public evidence first, then use account mapping after signature to confirm it.

Account mapping means comparing your CRM with a partner's to find the customers and prospects you share. Most guides on how to find channel partners for SaaS go straight to it. Forecastable, for example, recommends overlap reports from data-sharing tools. That works for partners you already have. With a firm you are trying to recruit, it puts the ask ("share your customer list") ahead of the reason ("here is proof we already share customers").

The leading guides also disagree on whether logos count. Forecastable says to "source candidates against the profile, not against logo lists." Partner.io says to "look where those partners already work with your customers: the agencies, consultants and software vendors your customers use," and names "their website, case studies, client logos" as places to check. Each is right about something different. This is how I reconcile them:

• The ideal partner profile is the gate. A firm that fails any trait is out, however many logos it shows.
• Public evidence is the score. Among firms that pass the gate, counted overlap sets the order.
• Account mapping is the audit. After a firm signs, run account mapping to test your estimate, then adjust your evidence weights for the next wave.

One limit, stated plainly: no published sample measures what share of agency sites show client logos, or how often those logos match a vendor's customer list. Treat your first ten firms as your own sample. Record how many showed logos and how many of those matched your customers, then use that hit rate to decide how far down the candidate list to research.

Which public signals prove a firm serves your customers

Four public signals carry overlap evidence: client logos and case studies, listings and tiers in adjacent vendors' partner directories, certified-staff counts, and service-line job posts. Each one proves less than it seems to, and each has its own typical way of misleading you.

How to read an adjacent-vendor tier

An adjacent vendor is a software company whose product your customers already run alongside yours. Its partner directory lists the firms that implement it. Tiers in those directories are public, structured, and comparable, which makes them tempting. They also measure the wrong thing for an implementation partner.

GRR is gross revenue retention: the share of recurring revenue kept from existing customers, before expansion. Read the table with three facts in mind:

• Zoho weights revenue heavily. Its 1,000-point score allocates up to 500 points to Revenue, 300 to Customer Success, 200 to Market Readiness, and 100 to Zoho Engagement. Authorized is the minimum tier for a directory listing, so an unlisted Zoho implementer may simply be new.
• HubSpot points track licence MRR. HubSpot awards points per $100 of MRR. An older HubSpot guide equates 2,100 sourced points with $42,000 MRR, so on that basis Gold's 110 sourced points come to roughly $2,200 of sourced monthly revenue. Two published sources also disagree on the Elite bar (2,100 / 9,000 against 2,750 / 11,000). Thresholds move between windows, so always read the live page.
• Salesforce reset its signal. Secondary sources report that Salesforce cut its consulting program from four tiers (Base, Ridge, Crest, Summit) to two (Select and Summit), and replaced 170 legacy certification badges with 28 competencies. Tier data from before the change is unreliable for scoring.

The consequence: tiers favour resellers over implementers. A delivery-heavy boutique can sit in a low tier while serving many of your customers well. Use tier as evidence of selling capacity and never as evidence of delivery quality.

How to use service-line job posts

A service-line job post is an opening for a role that delivers a specific practice, such as a RevOps consultant or a finance transformation lead. It is the freshest public signal you have: firms hire for practices they are growing, and the posting often names who the role reports to. Searching job boards one firm at a time is the slowest part of evidence collection. In Refolk you can ask in plain English for consultancies hiring a given role in a given market and get the people who run that practice back from LinkedIn, GitHub, and the open web.

How to check persona overlap, not just account overlap

Persona overlap means the firm works for the same buyer function that signs your deals, not just for the same companies. You can check it in three public places: the functions named on the firm's service pages, the titles of client contacts quoted in its case studies, and the roles it is hiring.

A shared customer is a large organisation with many buyers. An agency that runs paid media for a customer's marketing team shares that account with you. If finance buys your product, though, the agency has no seat in your deal. Kiflo lists "treating every overlap as equally valuable" and "assuming a customer relationship is automatically strong" among account-mapping failure modes. Forecastable names "presence at the buying committee" as a trait that predicts production. Neither publishes a way to check it from outside the firm, and no source documents how often account overlap without persona overlap fails. The method below is mine and has not been validated. Treat its grades as an input to ranking, not a verdict.

For each firm, write down three lists:

1. Service pages. Which function does each service line name as its audience? "For finance leaders" counts. "For growing businesses" does not.
2. Case-study contacts. Which job titles are quoted or named? A quote from a controller shows the firm sits with finance.
3. Hiring. Which roles is the firm posting for its service lines? A firm hiring RevOps consultants sells to revenue operations.

Then grade the firm:

• Same function: at least two of the three lists name the function that signs your deals.
• Adjacent function: the lists name a function that influences your deal but does not sign it.
• Different function: none of the lists names your buyer or anyone who influences the deal.

A same-function grade tells you the firm is already in the rooms your AE needs to be in. It misleads when a service page names nearly every function, so every vendor matches. In that case, ignore the service-page list and grade from case studies and hiring alone.

A shared customer is not a shared buyer. Overlap counts only when the firm sits with the person who signs.

The partner recruitment process, step by step

The process has seven steps, run in order, each with an owner and a done-state. Skipping a done-state is how a ranked list slides back into a brainstorm.

How the two days split

No published source times these steps, so treat this split as a planning target, not a benchmark:

• Day one: steps 1 to 4. Evidence collection in step 4 takes the longest, so start it as soon as step 3 turns up its first firms rather than waiting for the full candidate list.
• Day two: steps 5 to 7. Once overlap scores exist, the AE's persona check and the partnerships lead's ranking can run in parallel. Naming owners and sizing the cohort close out the day.
• If day one ends with fewer than 20 firms scored, widen the seed list in step 2. Do not loosen the profile in step 1.

How to score, rank, and name the owner at each firm

Score each firm by counting the distinct best-fit customers it shows evidence for, weighted by how strong that evidence is. Then rank on overlap strength, strategic fit, and reachability, and cut the list to 20 to 40 firms. Every firm that makes the cut leaves this stage with a named practice lead, a next action, and a deadline.

Ideal partner profile scoring works best when it is boring: count, don't judge. Count each best-fit customer once, at the strongest evidence you found for it, so a firm that shows one customer three different ways scores one customer, not three. Forecastable ranks on "overlap strength, strategic fit, and reachability." Map those onto the sheet like this:

• Overlap strength is your weighted count.
• Strategic fit is the persona grade plus motion fit.
• Reachability is whether you have found a named lead, and whether you have any warm path to them.

Forecastable sizes the working list at 20 to 40 partners.

Firm:
Website:
Motion (resell / refer / co-sell / build):
Competing service line? (yes = stop here):

OVERLAP EVIDENCE (count each best-fit customer once, at its strongest evidence)
Dated case study naming a best-fit customer, with named client contact: 3 points each
Undated case study, or case study without a named contact: 2 points each
Logo only: 1 point each
Best-fit customers evidenced (count):
Overlap score:
Sources (link and date for every point):

CAPACITY EVIDENCE
Adjacent-vendor directory and tier (read in live directory on [date]):
Certified-staff count (source, date):
Open service-line job posts (titles, dates):

PERSONA OVERLAP
Functions named on service pages:
Titles quoted in case studies:
Roles being hired:
Grade (same / adjacent / different) and one-line reason:

RANKING
Strategic fit (high / medium / low) and reason:
Reachability (named lead found? warm path?):
Rank:

OWNER
Practice lead name and title:
Evidence they own the practice:
Next action:
Deadline (inside 90 days):

Who to name at each firm

Who owns vendor relationships inside service firms is not established publicly. Forecastable's reseller example names a champion "at the partner's regional director level," but that is one illustration, not a measurement. So name the person the evidence points to, and write down why:

• The contact named on the firm's adjacent-vendor directory listing.
• The hiring manager or reporting line on a service-line job post.
• The practice lead quoted or credited in the case studies you counted.
• At small firms with no visible practice lead, the founder.

Kiflo's warning holds on the recruiting side too: "an account without a named owner on both sides tends to stall." A firm with no named lead drops to the bottom on reachability. It stays on the list.

This is where the search effort peaks: 20 to 40 firms, each needing the one person who runs the relevant practice. In Refolk you describe the firm and the role in one sentence and get the people back.

How big the recruit cohort should be

The recruit cohort is the number of firms you need to sign to end up with the productive partners you want: the target number of productive partners divided by the activation rate you can honestly expect. For five productive partners in a typical managed program, which activates 30 to 50% of recruits, that is 10 to 17 signed firms.

Introw defines activation as a partner's "first deal registration or first closed-won opportunity within 90 days." Unifyr's Channel Atlas puts typical activation for newly recruited partners at 30 to 50%, and below 20% for programs that do not actively manage activation.

The cohort column is derived: 5 divided by the rate, rounded up. The first three rows come from Unifyr, the affiliate rows from Track360.

Pick the row that matches how you will actually run the program. If no one owns partner onboarding week to week, use the unmanaged row and plan for 26 or more. No source splits activation cleanly by reseller, services, and technology partners. Introw notes only that referral partners can activate quickly, while enterprise system integrators often need more training. If your list leans toward large integrators, plan from the low end of the managed range.

Track360's affiliate data shows direct outreach activating at roughly two to three times the rate of open registration. The direction probably holds for targeted services recruiting, but that is not established for services partners, so do not borrow the figure.

The 90-day deadline on every next action is there for a reason. Introw reports that partners with a first deal inside 90 days are three to four times more likely to still be active after one year. Every first action should aim at a joint opportunity inside that window.

What payoff to promise the business

Crossbeam's figures only cover companies already account mapping on its network, so selection bias is likely. Allbound reports that deals are 53% more likely to close with a partner involved, but use the conservative number in any forecast you hand to finance.

How partner sourcing goes wrong

Most of the common failures make a firm look like a stronger partner than it is: inflated logo counts, stale or misread tiers, shared accounts with the wrong buyer. The rest break the cohort maths or stall the list after handoff. Each one has a cheap check.

1. Logo walls. "A logo wall is a vanity metric," and old or one-off clients inflate the overlap count. Check: score a logo at 1 point unless a dated case study with a named client contact backs it up.
2. Stale tier labels. Any page still showing Salesforce Ridge or Crest predates the two-tier model. Some sources reverse the old order, listing "Base (up to 249 points), Crest (over 250 points), Ridge (over 500 points)." Check: read the tier in the live vendor directory and record the date.
3. Reading tier as delivery quality. As one agency comparison puts it, "tier reflects sold monthly recurring revenue and retention, not the quality of any given project." Check: use tier only as selling capacity, and never mark a firm down for a low tier when its dated case studies are strong.
4. The competing partner. A firm whose services overlap your product will keep the business instead of referring it. Check: if a customer could buy either the firm's service or your product to solve the same problem, mark the firm as competing and remove it.
5. Persona mismatch. The accounts are shared but the buyer is different, which is the "assuming a customer relationship is automatically strong" failure Kiflo flags. Check: the persona grade, built from service pages, case-study contacts, and hiring.
6. Activation-benchmark mismatch. Mixing affiliate activation (a first click within 30 days) with services activation (a first deal within 90 days) makes the cohort look smaller than it needs to be. Check: the cohort note names a services benchmark row.
7. Survivorship in win-rate data. Crossbeam's lift "only applies to an organization's connected ecosystem." Check: forecast from the 1-5 partner figure, not the network average.
8. Recruiting before ranking. Signing everyone who says yes gives you the sixty-partners, six-producers outcome. Check: nobody gets a first message until the list is ranked and cut.
9. No named owner. A firm with no named contact stalls after handoff. Check: every row has a name, the evidence for that name, a next action, and a deadline.

Before handoff, and how to keep the list current

The list is ready for the partnerships lead when every firm passes the profile, carries a traceable overlap score and a persona grade, and has a named owner with a dated next action. After handoff, re-check tiers and job posts on a schedule, because both go out of date.

The first message should lead with the overlap you proved, not with a request for data.

Subject: [Customer A], [Customer B] and your [practice name] practice

Hi [first name],

I came across your case study with [Customer A] and your [tier] listing in [adjacent vendor]'s partner directory. [Customer A] and [Customer B] both run [your product] as well, and at both the [buyer function] team owns it - the same team your [service line] work is built for.

I am putting together a small cohort of partners who already implement for that buyer. I am not asking for data or a commitment, just a short call to compare where our customer lists overlap and whether a first joint deal in the next 90 days is realistic.

Would [day] or [day] work?

[Name]
[Title], [Company]

To keep the list current:

• Re-read tiers whenever vendors re-evaluate. Secondary sources report that Salesforce now evaluates partners quarterly. HubSpot publishes thresholds per window, and the published HubSpot figures already disagree on the Elite bar.
• Re-run job-post searches before each recruiting wave. A practice that has stopped hiring may have stalled. Save the Refolk query you used to find practice owners, and the refresh becomes a re-run rather than a fresh search.
• Audit against account mapping after each cohort signs. With the partners who agree, compare their true overlap against your public-evidence score, then adjust the weights on the scoring sheet.
• Record your logo hit rate and signing rate every wave. No published source gives you these two numbers, and after two waves they will tell you more than any benchmark here.