Scoring an Account's Fit Before It Earns Rep Time

This is a rubric for one decision made many times: is a single named company worth a rep's time before it enters the pipeline? It is written for founders selling their own product, account executives, SDR leads, and partnerships teams who have a company in front of them and no CRM enrichment or intent subscription feeding a score. You will finish able to grade that company on structural fit and buying window from public evidence, then assign a defensible tier with the one reason that drove the call.

Most account-scoring guides assume a stack: enrichment feeds firmographics, an intent vendor feeds timing, and a model spits out a number. This one grades a single account by hand from public, verifiable evidence, and it keeps fit and window as two numbers on purpose so they never average into something misleading.

Why fit and window must stay two separate numbers

Fit describes what a company is; the buying window describes what it is doing right now. Best-practice scoring keeps them as two numbers rather than blending them, because they decay on completely different clocks.

Firmographics barely move month to month. A funding window is strongest 2 to 4 weeks after the announcement, a new VP hire opens a 30 to 90 day window, and pricing-page interest fades within 5 to 10 days. Blend a stable fit read with a volatile timing read and the composite is wrong within a fortnight: a structurally perfect but budget-frozen account scores hot, and a rep burns hours on a company that will not buy for a year.

The cost of getting this wrong is not abstract. At any moment only about 5 to 10 percent of your ICP accounts are actually in-market. If you score fit-high accounts as pursue-now, you flood reps with structurally-good, timing-wrong work and dilute the queue that matters.

The payoff for fit discipline is measurable on the other side. ICP-fit customers post net revenue retention of 110 to 130 percent, against 70 to 90 percent for off-ICP customers. Organizations with a strong ICP achieve 68 percent higher account win rates, per TOPO (now Gartner). And well-defined ICPs correlate with 24 percent lower churn and 33 percent higher expansion revenue. Fit is not paperwork; it is the difference between an account that expands and one that leaves.

The order of operations: disqualify, then fit, then window

Run the account through three gates in order: a disqualifier gate, a structural fit score, and a separate buying-window score. The order is not cosmetic. You do not score timing on a company that was never a real fit, and you do not score fit on a company that fails a hard exclusion.

Sources are near-unanimous that fit is the first qualifying gate. ICP fit must be qualified before any deal-qualification framework like MEDDPICC or SPICED, because no framework compensates for fundamental organizational misalignment. If the account fails the ICP test, the opportunity should never enter the forecast. Intent then works as a prioritization layer on top of a solid fit filter: first filter for fit, then among good-fit accounts use timing to choose who gets attention now.

One dissent worth naming: a minority approach (the 6QA model) requires fit, intent, and engagement together before an account earns a "qualified" status, rather than gating fit first. If your motion depends on engagement signals to qualify, adapt the sequence. For cold, pre-pipeline triage from public evidence, gate fit first.

The disqualifier gate: what stops a score cold

Before you score anything, run the account against an anti-ICP exclusion list. A negative ICP is a defined set of exclusions applied before scoring begins, and it is worth more than any positive weight you can add.

The mechanism is attention concentration, not better selling. One reported case at a Series C fintech implementing negative scoring cut lead volume by 40 percent yet lifted win rates by 22 percent. Treat that as illustrative rather than a benchmark, but the direction is consistent with the tier data: concentrated attention on fewer, better accounts converts harder.

Gartner's test is the sharpest way to know your list is finished: "What defines an account we can't sell to?" If your team cannot answer in one sentence, the ICP is not done.

Scoring structural fit: what each signal proves and how it lies

Structural fit is a 0 to 100 total built from three inputs: firmographics define the pool, technographics rank it, and authority confirms someone can actually buy. Each signal proves something specific, and each has a way of lying that you check before you trust it.

Firmographics indicate organizational fit but do not guarantee success. Actual results depend on use-case alignment, stakeholder engagement, and value realization, so read them as a pool filter, not a verdict. Technographics answer what a company runs and reveal product fit, integration needs, and competitive displacement openings. They are gathered from public signals: detections of code and tracking tags on a company's site, job postings that name specific tools, and app-marketplace and DNS records.

The technographic lie is the dangerous one. A detection going dark is worth investigating, but not every gap means a tool was removed. A script change or a recrawl gap produces the same result. Corroborate a dark detection against a live job post or the site itself before you treat it as a displacement opening.

Authority deserves its own caution because buyer pools are geographically lopsided. In Refolk's index of professional profiles, the United States holds about 8,509 senior sales-leadership contacts (VP of Sales, CRO, Head of Sales) against about 5,092 in the United Kingdom - the US pool is roughly 1.67 times larger. A US-calibrated authority threshold will over-disqualify UK accounts where the buyer pool is genuinely thinner.

Some buyer roles are scarce enough to be a signal in themselves. In Refolk's index, US Revenue Operations leaders number about 384 against 8,509 sales leaders - just 4.5 percent of the sales-leadership pool. A new RevOps hire is therefore a scarcer, higher-signal leadership change than a routine sales-VP move.

Scoring the buying window separately

The window score answers a different question: is now the moment? Grade it from public triggers, keep it apart from fit, and remember that a single signal rarely justifies action.

The window multiplier dwarfs everything else you control. Trigger-referencing emails reply at about 18 percent against a 3.43 percent baseline - a gap of more than five times that comes from timing, not wording. Marginal rep hours convert where the window is open, which is exactly why the window earns its own number.

Not all signals are equal, and each has a length. Score the stack, not the spark.

Read the table as a hierarchy. A former champion arriving in a buying role converts at roughly three times cold outreach, and close rates run 114 percent higher when a previous champion sits in the buying group. Newly hired execs spend about 70 percent of their budget in the first 100 days, so the 90-day window after a decision-maker starts is the most reliable period to move. Funding-round outreach within 48 hours converts at roughly four times the baseline, with a 60 to 90 day evaluation window behind it.

The bottom row is the trap. Job postings alone add only about 7 percent. Teams that over-index on postings are chasing noise. Never fire on a single signal if you can wait for a stack to form.

The prompt below turns a window read into a list of named accounts you can score. Ask for the trigger and the fit filter together, and you get companies where both a signal and structural fit are already present.

Finding one such account by hand means reading funding databases, job boards, and technographic detectors and reconciling them. Refolk does that reconciliation from a plain-English ask, returning the companies where the trigger and the fit filter both hold so your scoring hours go to accounts worth scoring.

The five-times reply gap between trigger emails and cold ones is timing, not copy. Score the window.

Run the rubric: from named company to logged tier

Here is the procedure end to end. It moves one named company through the disqualifier gate, a fit total, a separate window score, and a tier with a driving reason. Budget about 45 minutes the first few times; it compresses with practice.

The tier cutoffs follow the dominant convention on a 0 to 100 fit scale. Route by tier and SLA, and let the window score decide sequencing within Tier A.

The tiering earns its keep in the outcome data: top teams see Tier A win rates 1.5 to 2 times higher than Tier B, with 15 to 20 percent shorter cycle times. That gap only survives if the bar is high enough to keep Tier A small.

To keep tiers from oscillating as scores drift, use hysteresis: promote an account into Tier A at a composite of 85, but do not demote until it drops below 70. And apply time decay so a stale score does not masquerade as fresh: a common convention reduces the score 25 percent monthly without activity.

How this goes wrong: the failure modes

The failure modes are where a scoring standard earns its place, because a rubric that overclaims is worse than none. Each entry below is a false positive and the check that catches it.

• Averaging fit and window into one number. A structurally perfect but budget-frozen account scores hot. Check: keep two separate scores.
• Firing on a single signal. A lone job post is usually a backfill. The real signal is a surge, multiple roles in the same department in the same window at an ICP-fit company. Check: require a stack before you move.
• Trusting a dark technographic detection. A recrawl gap mimics a tool removal. Check: corroborate with a job post or the live site before calling it a displacement opening.
• Building the ICP from closed-won only. Accounts that closed but churned within a year are anti-signals, not proof of fit. Check: build the ICP from retained-and-expanded accounts and overlay NRR and churn on the sample.
• Setting the bar too low. If everything scores into Tier A, the 1.5 to 2x close-rate advantage disappears. Check: the Tier A queue must fit your response SLA.
• Using too many attributes. With 15 attributes every prospect kind of fits and kind of does not, so the ICP stops pointing at the best opportunities. Check: won deals should cluster in A and B, lost deals in C.
• Leaving a stale hot score in place. An account that scored 85 six months ago and has not engaged since is not Tier A anymore. Check: apply time decay every month.
• Contacting the wrong person after a signal fires. A funding signal points to the budget owner; a leadership change points to the new leader. Check: match the persona you contact to the signal that fired.

Keep it current: recalibrate against outcomes

A fit rubric is only as good as its last recalibration, so treat scoring as a loop, not a one-time build. Quarterly, compare win rates, deal sizes, and cycle times across tiers, and adjust criteria or weights if Tier A is not converting meaningfully better than Tier B.

Two mechanical habits keep the day-to-day scores honest between recalibrations. Apply the 25 percent monthly decay so idle accounts fall out of the hot queue on their own. And use hysteresis - promote at 85, demote below 70 - so accounts near a boundary do not flip tiers on noise. A healthy account-to-opportunity conversion for worked tiers runs roughly 60 to 80 percent; if a tier falls well below that, its criteria are letting the wrong accounts in.

Before you call any single account scored, run the checklist.

Keep the rubric small, keep the two numbers apart, and let the outcome data prune your criteria. A scoring standard that names the accounts it cannot sell to, and stops cold when one appears, will send reps at fewer companies and win more of them.