The Deal Qualification Score: Commit, Work the Gaps, or Walk

You have ten open deals and a forecast cycle in two days. Your job is to say, for each one, whether it belongs in the commit forecast, needs more work before it earns that status, or should be disqualified and taken off the sheet. This guide is for founders selling their own product, account executives, SDR leads, and partnerships teams who have to make that call and defend it to a manager. It gives you one evidence scale across eight checks, a way to verify the weak ones against public signals, and a commit cutoff two reviewers would apply the same way.

Most published scorecards map the buying committee, grade one champion, and pin the economic buyer. Few of them turn the whole opportunity into a forecast decision with a threshold a second reviewer could reproduce. That is the gap here. I fix a single evidence scale per check, tie each weak check to a specific public signal you can go verify, and set a commit cutoff so that two people grading the same deal land in the same tier.

What the eight checks are and why they decide the forecast

The eight MEDDPICC checks are Metrics, Economic Buyer, Decision Criteria, Decision Process, Paper Process, Identify Pain, Champion, and Competition. Grading all eight on one scale is what turns a pile of impressions into a forecast decision you can audit.

Each check asks a different question about whether the deal is real. Metrics asks whether the buyer has quantified the value. Economic Buyer asks whether you have met the person who signs. Decision Criteria asks whether you know the standard you are being judged against. Decision Process asks whether you know the steps to signature. Paper Process asks whether you know the legal and security path. Identify Pain asks whether the problem is urgent enough to fund. Champion asks whether someone inside will sell for you when you are not in the room. Competition asks who else is being considered, including the status quo.

The reason to score all eight rather than eyeball the deal is that the checks fail independently. A deal can have a loud champion and a quantified metric and still die because nobody mapped the paper process until a security questionnaire landed in week nine. The scorecard forces you to look at the check you would rather not look at.

In Refolk's index of professional profiles, 1,536 US account executives list MEDDPICC as a skill and 271 list MEDDIC, against a US pool of 263,961 account-executive and enterprise-account-executive profiles. Under one percent advertise the discipline. The skill is scarce, which is exactly why a written standard beats a shared assumption: you cannot rely on everyone already scoring the same way.

The evidence scale: what each rung has to prove

Score every check on evidence, not confidence. The rung you pick is a claim about what the buyer said or did, never an adjective about how the deal feels.

The scale that two reviewers can apply identically is the one where each rung carries a test. A 0-3 scale, well documented, runs like this: 0 is no information and the rep cannot say anything; 1 is an assumption the rep believes but no buyer stated; 2 is stated by the buyer, meaning a named person said it and the rep can quote them; 3 is confirmed and acted on, meaning it is written down, the buyer has agreed the write-up, and it has changed the deal plan. A 0-4 variant inserts a "tested with the buyer" rung below "documented": 0 unknown, 1 assumed, 2 stated, 3 tested, 4 documented.

The load-bearing principle is the same across every source: the score tracks evidence. If a rep feels they have a champion but cannot describe an action that champion has taken, the score is low. "Strong champion" is not a score. "Introduced me to the VP of Finance on the 14th and forwarded the business case" is a score.

Which scale you choose matters far less than everyone using the same one. A champion scored 4 by one rep and 1 by another is a pipeline-review disaster, because the number no longer means anything when it rolls up.

The commit threshold: where to draw the line

Set the commit floor at roughly 70% to 80% of the maximum, with a rule that no single check sits at its lowest rung. Published scorecards cluster tightly in this band, and the floor only works if you also gate on the worst check, not just the total.

A raw total can hide a fatal gap. A deal can clear 75% on the strength of six strong checks while the Economic Buyer sits at 1. That is why the better-documented thresholds layer a per-check floor on top of the percentage. The MEDDIC template rule is 13 of 18 on a 1-3 scale with no letter at a 1, and specifically no red on Economic Buyer or Champion. The weflow variant drops the total entirely: every deal in the commit forecast must have zero reds and no more than two yellows.

Pick one and write it down. The Kalungi matrix is the cleanest three-way split: 75% and above is Green and commits, 50% to 74% is Yellow and needs coaching, and below 50% is Red and should be disqualified. Saber's four-tier version maps the same idea onto the forecast language managers actually use: 80% to 100% commits, 60% to 79% is best case, 40% to 59% is pipeline, and below 40% is unqualified and should not be forecast at all.

Why the Economic Buyer check does most of the predictive work

If you weight one check above the others, weight Economic Buyer access. Its absence is one of the strongest predictors of slip or no-decision in qualification frameworks, and the numbers isolating it are the most citable in the field.

The mechanism is simple. The economic buyer applies criteria the seller never saw and can veto the deal at approval. You can run a flawless process with a friendly champion and still lose because the person who signs was never sold. The data lines up behind the mechanism. Deals where the seller never met the economic buyer win about 12% of the time versus 54% with direct access. iSEEit's research found the economic buyer had not been met in 6 of 7 lost or slipped deals, while a direct EB meeting had occurred in 5 of 6 won deals; when the economic buyer supports the project and timeline, close rates rise above 80%. Gartner research puts B2B purchases at 6 to 10 stakeholders, with verified EB engagement roughly three times more likely to close.

There is a related signal worth tracking that sits outside the eight checks entirely. The best real-time predictor that a committed deal will slip is a rep changing the close date, and it usually arrives before any stage change or score drop. Count close-date pushes as a field. The second push is the one to act on. Research from Ebsta and Pavilion's 2025 B2B Sales Benchmarks found early decision-maker involvement boosts win rates by 55% while delayed engagement reduces them by 113%, which is the same story told from the front of the deal rather than the end.

How to score a deal: the eight-check procedure

Run this once a week per active deal and at every event trigger. It takes a rep ten to fifteen minutes once the notes habit is in place.

The step that separates a scorecard from theater is the fifth one. A deal review that walks the highest scores is theater. Walk the 0s and 1s, ask what next action would move them, and assign an owner. The point of scoring is to surface gaps, not to congratulate the deal on its strengths. MEDDICC's "100-hour rule" frames the budget plainly: roughly one hour a week of deal reviews across about 44 working weeks is what the discipline costs.

Verifying the weak checks against public signals

When a check scores 0 or 1, you have a choice: get the buyer to raise it, or corroborate it from outside the deal. Several checks have a public signal you can verify before the next call, and knowing which ones do keeps you honest about the rest.

A real buying signal has a verifiable source: a press release, a job posting, an SEC filing, or a product-review visit. A new executive appointment informs the Economic Buyer and Champion picture. Hiring surges and budget approvals point at Decision Criteria and Decision Process. For the Paper Process, the strongest close-proximity signals are procurement or legal involvement: requests for security questionnaires, SOC 2 reports, or redlined contracts. The free verification stack is enough for most of this: Google Alerts for leadership changes, LinkedIn notifications for job changes, SEC EDGAR for public-company filings, Crunchbase for funding alerts, and job-board monitoring.

Be honest about the limit. For several checks the only public signal is inferred from job titles, which is weak. A buying-committee map built from LinkedIn titles is a starting map for multi-threading, not a statement about how the company actually runs. Confirm a role against an action, not a headline. The Champion check in particular is behavioral, not public: if your champion is unwilling or unable to introduce you to the economic buyer, that itself is a signal they are not a true champion.

Re-verifying the economic buyer is the highest-leverage version of this work, because the EB moves. A new finance leader resets the approval criteria on every deal in their patch. Refolk turns that from a manual LinkedIn sweep into a single plain-English request, which matters when the EB check is the one carrying most of your forecast risk.

How this goes wrong: failure modes and false positives

Most bad forecasts come from a small set of predictable errors, and nearly all of them are a green score that lies. Learn to spot the false positive for each one, because the scorecard is only as good as your suspicion of it.

The errors behind that checklist are worth naming one by one.

Scored once, never re-scored. A green scorecard frozen at first contact is worse than no scorecard, because it manufactures confidence. A green field in March can be red by May. The check is dated notes and a changed number.

Scoring from confidence, not evidence. "Strong champion" with no action named is the classic false positive. Demand a quote or an action the buyer took.

Skipping the Economic Buyer because the champion is easy. The most expensive error. The deal looks committed on a friendly champion's word, but nobody confirmed the signer saw the business case. Ask the one question: has the signer seen it.

Mistaking an enthusiastic contact for the budget holder. Mistaking an enthusiastic VP for the budget holder is one of the most common reasons a well-qualified deal slips at the end of the quarter. The check is veto power over budget, not volume of enthusiasm.

Treating title-inferred committee maps as verified. An org chart built from LinkedIn titles is a hypothesis. Confirm each role against an action.

Treating a close-date push as neutral. It is the single best slip predictor. Count it.

Paper Process left blank until it is too late. Surface it by week three. A late security review is a classic end-of-cycle loss.

A green scorecard frozen at first contact is worse than no scorecard, because it manufactures confidence.

There is also an error in the other direction. Over-sandbagging shows up as commit accuracy that is consistently 100%. If every committed deal closes, deals that belong in commit are stuck in best case and your forecast understates the business. The target is a band, not a ceiling.

Keeping the standard current

The scorecard is a living instrument, so the maintenance work is measuring whether your threshold still predicts. Track conversion by tier every quarter and adjust the cutoff against the accuracy band, because a threshold that worked last quarter can drift as the market or the team changes.

World-class commit accuracy is 90% to 95% of committed deals closing. Consistently below 80% means your commit criteria are not stringent enough and you are letting weak deals in. Consistently above 95% means reps are sandbagging and the real forecast is higher than the sheet. Commit accuracy is a two-sided error, and the only way to know which side you are on is to measure conversion by tier and read it against that band.

Two scheduling debates are worth settling for your team. The first is cadence: some sources re-score at milestones only, others insist on fixed weekly cadence regardless of milestones. Run both. Weekly catches slow decay; triggers catch sudden change. The second is the gate: some teams gate on percentage total, others on zero reds and no more than two yellows. The per-check gate is safer because it cannot be fooled by a high average hiding a fatal zero.

The audit skill itself is scarce, which is the last reason to write your standard down rather than carry it in your head.

The US account-executive pool in Refolk's index is about 13.4 times the size of the UK pool, yet across both markets the named qualification skills barely register.

MEDDPICC is listed 5.7 times more often than MEDDIC among US profiles, and the top employers of MEDDPICC-skilled profiles in the sample skew toward cybersecurity and infrastructure vendors such as SailPoint, Cohesity, OPSWAT, Amperity, and SUSE. If you want to hire reps who already grade deals this way, or benchmark your bar against teams that do, Refolk lets you find enterprise account executives who list MEDDPICC and work at cybersecurity vendors in one request. The discipline is rarer than the roles, so writing it down and training to it is most of the advantage.