Ask a supplier what their win rate is and you get a number somebody estimated once
Not because anybody is careless. The bids live in a spreadsheet and the outcomes live in a federal database, and joining a year of pursuits to their results by hand is a day of work nobody has. Both halves are already here: you marked a notice submitted, and the government later published who won it.
concentration 4,072: this is a displacement, not a bid counted on resolved identity, or one firm reads as five
The outcome is found. The decision stays yours.
Every pursuit you marked submitted is checked against the award published on that notice. Where the award is recorded against your own resolved company, it suggests a win. Where it went to somebody else, it suggests a loss, and names them.
It never marks anything for you, for two reasons that both matter. A notice with no award published yet is not a loss, it is a decision that has not been made, and counting it as one makes every win rate too low and every pipeline look dead. And a company match is a judgement, not a fact, so a product should not quietly rewrite your own history with it.
- Suggested from resolved company identity, not from two spellings agreeing
- The winner, the award number, the date and the amount as published
- Undecided pursuits shown as undecided rather than counted as losses
- One click to confirm won, lost or no bid
a score with no reasons is a horoscope
Where you actually win, and where you never have
A win rate over everything is one number. The same rate broken down by buying office and by code is a plan: these three offices you win at, this one you have bid nine times and never taken, and that is a conversation to have rather than a tenth bid to write.
It is counted only over pursuits that were settled. Everything still open is excluded and reported separately, because a bid nobody has answered is not a loss, and a rate that drifts down every week something sits unanswered is a rate nobody should act on.
- By buying office and by NAICS code
- By count and separately by money, which are different questions
- Open pursuits excluded from the rate and reported on their own
- No bids kept out of it, because deciding not to bid is its own decision
the filter nobody else offers is the last one
A loss recorded as a word is a tally mark
Settling a pursuit asks one more question: why you think it went the other way. From a short fixed list, because a reason nobody can count is not one, and a list of thirty cannot be chosen from honestly at the moment somebody is annoyed about losing.
Then it is counted. Most companies that lose a few in a row have a theory about why, and the count either confirms it or replaces it with something they can act on. Most of what a company loses, it usually loses on one thing.
- Price, past performance, technical score, the incumbent, found non-responsive, ran out of time, cancelled
- And "we never found out", which is a finding about how the company runs rather than about the bid
- A debrief is something you can ask for on a negotiated buy, and the window to ask is short
- The losses with no reason on them are counted too, and named
the gap is the source publishing late, so the page says which feed it heard from and when
Are you actually expensive, or does it only feel that way
Record what you bid when you settle a pursuit and the other half is already on the notice: what it went for, filled in from the published award rather than asked for. Nothing in this market compares those two numbers, and they have been sitting under identical keys the whole time.
The answer is a median rather than an average, because one bid at triple the eventual price is a mis-scoped estimate rather than a pricing habit and it moves an average enough to say the opposite of the truth. And it says nothing at all until four settled bids carry both numbers, because a pricing verdict from two is how a company talks itself into cutting a margin it did not need to cut.
- How far above or below the eventual price your bids land, as a median
- The same split between the ones you won and the ones you lost, which is the comparison that says whether price is deciding them
- Nothing claimed under four settled bids, and it says why
- Your own history only, and never anybody else's
the shape of the distribution, not a point on it
Winning a bigger share of the money than of the count means something
It means the large pursuits are going your way and the small ones are not, and those small ones may be costing more to bid than they return. The reverse means the big ones are going elsewhere, which is usually a past performance problem rather than a price one. A single blended win rate hides both of these completely, which is why the two are counted separately and the difference between them is read out for you.
A rate on three bids is a direction, not a number
Eight settled pursuits is the point at which this stops saying so on every screen. Below that the figure is shown with the count beside it and a line saying to treat it as a direction, because a 33 per cent win rate printed against one win in three is exactly the kind of figure that ends up in a capability statement and then has to be defended in a room.
Where it is in the app
Win and loss, under Contract in the menu. Contract is the part of LastAwarded that runs the award, one of the eight parts.
Related
- Pipeline and calendarWhere a pursuit is marked submitted.
- Contract ledger NewWhat happens to the ones you win.
- Bid worksheetWhat you priced them at.
- Incumbent profilesThe companies taking the ones you lose.
See who was last awarded in your NAICS
Every code, every buying office, every vendor and every recompete, organized from the federal record and read nightly. Open it with a plan.