How to Measure Competitive Deal Coverage

A competitive win rate tells you how the deals you know about went. It cannot tell you whether the team was equipped for them, and it quietly assumes that every deal worth counting made it into the count. Coverage is the metric that tests that assumption.

It answers a narrower question than most programme metrics: of the deals where a competitor was in the room, how many had the relevant intelligence in front of the rep while the decision was still open? Not how good the intelligence was. Not whether it was used. Whether it arrived.

Both halves come from a single CRM export of closed deals. The numerator needs one more field than most pipelines carry, which is the date the briefing was delivered, and that field is the entire implementation cost of this metric.

Note what the example does not do. It does not average across reps, and it does not report a trend from a single quarter. Coverage is lumpy by account team and by rival, and the useful cut is almost always by competitor: a function at 41% overall is frequently at 80% on the rival everyone worries about and near zero on the two that are quietly winning deals.

Strip the timing requirement out and this measurement collapses into something worthless within a quarter. Every competitive deal eventually gets analysed, because losing to a competitor generates a post-mortem. If analysis counts, coverage converges on 100% while the sales team experiences no change whatsoever.

What counts as briefed

Three conditions, all of them: specific to that deal, delivered to the people working it, dated before the close. Applied to what a competitive function actually sends, that rules more in and out than teams expect.

The digest row is the one people argue about, and it is worth holding. Counting broadcasts is how a programme reports 90% coverage in a quarter where nothing reached anyone in particular.

Dating the briefing when nobody logged it

Most teams start without the delivery date and try to reconstruct it. The reconstruction is usually possible for one quarter and never accurate beyond that, so treat the first measured period as a rough baseline and start dating deliveries from that point. A coverage series that begins with one honest estimate and then improves in quality is more useful than one that quietly back-fills.

This is the argument for measuring coverage first, and it is not an argument about coverage’s own importance. Every competitive metric drawn from the pipeline shares one denominator: the deals somebody recorded as competitive. Coverage is the only number that tells you how complete that record is.

Below roughly half, the recorded competitive deals are not a sample of your competitive deals. They are a sample of the deals where a rep both noticed a competitor and bothered to write it down, which correlates with deal size, rep seniority and how threatening the competitor felt. A competitive win rate computed on that population will read higher than reality, because the deals nobody logged are disproportionately the ones that were lost early and quietly.

The limitation to say out loud

Coverage cannot see the deals where a competitor was present and nobody recorded it. That makes every coverage figure an upper bound on the truth rather than an estimate of it, and it is why setting up the competitor field properly is a prerequisite for this metric rather than an improvement to it. A tidy 60% drawn from a pipeline where half the contested deals are unmarked is a worse number than a messy 40% drawn from one where they are not.

1. 1Export closed deals for the period with the competitor field, the close date and the owner.
2. 2Filter to deals naming a competitor. That is the denominator, and its size relative to total closed deals is worth writing down on its own: a pipeline reporting 8% competitive deals in a contested market has a recording problem, not a market position.
3. 3Join the briefing log and keep only deliveries dated before the close.
4. 4Divide, then immediately split by competitor and by owner. The aggregate is for the deck; the splits are what anybody can act on.
5. 5Report the uncovered deals by name. A list of nineteen deals that closed without a briefing starts a different conversation from the number 41%.

Where a connected CRM changes the work

The whole calculation is a CRM export, a filter and a division, which means it belongs in a spreadsheet and takes about ten minutes once the fields exist. Keep the export rather than just the result: a quarter of raw rows is what lets you re-cut coverage by competitor later without re-running anything.

A competitive intelligence platform connected to the CRM changes the numerator rather than the arithmetic. When a competitor is named on an opportunity, the relevant material can be attached to that deal automatically and dated as it happens, which removes the reconstruction problem described above and makes the briefing log a by-product instead of an admin task. The definition of coverage, and the decision about what counts as a briefing, stay yours.

Coverage fails at a specific moment, and it is not the moment anyone expects. It fails when a deal turns competitive between two forecast calls. Nobody is told, the rep handles the objection from memory, and the deal is closed by the time the competitive function learns a competitor was ever involved. The work existed; it was never routed.

The cost of that lag lands on exactly the wrong deals. A competitor appearing late in a cycle usually means the buyer is at the comparison stage, which is the point where a specific, current answer changes the outcome and a generic one does not.

Closing that gap is what competitive intelligence platforms connected to a CRM are for: watching for a competitor appearing on a live opportunity and putting current material in front of the people working it, without anybody filing a request. Flares works this way, keeping the underlying competitor material current so what gets routed is worth reading when it arrives.

What automation cannot do is decide what a briefing should say for a deal it has never seen, or notice a competitor nobody named. Coverage measures routing, and routing is the part a machine is genuinely good at. The judgement about which competitors matter, and what to tell a rep about them, does not move.