Tracking Your Bid Win Rate — and What It Actually Tells You

Ask a traffic control owner how busy the company is and you will get a confident answer. Ask what share of quotes turned into jobs last quarter, and by customer type, and the answer is usually an estimate — often a flattering one, because wins are memorable and quotes that went quiet are not.

Win rate is one of the few numbers that tells you something about pricing, sales and estimating capacity at once. It is also easy to measure badly. Here is how to make it mean something.

Define the denominator first

Win rate is wins divided by bids, and the argument is always about what counts as a bid.

A workable definition: a priced quote, sent to a customer who requested it, for a specific job or a rate sheet. That excludes casual phone conversations and the quotes you declined to submit. It includes bids you submitted and lost, bids that were cancelled by the customer, and bids that were never decided.

Then keep the categories honest:

• Won — you were selected.
• Lost — someone else was.
• No decision — the job was cancelled, deferred, or simply went silent.
• Withdrawn — you pulled it.

"No decision" matters. A company whose quotes mostly go silent has a different problem from one that loses on price, and merging the two hides it.

Record enough to learn something

The bare minimum per bid:

| Field | Why it matters | | --- | --- | | Customer and type | Public agency, prime contractor, utility, private | | Date sent | Drives response-time analysis | | Job type | Lane closure, flagging, device rental, special event | | Road type / speed | Complexity proxy | | Value | Lets you weight by dollars, not just count | | Prepared by | Estimator-level differences are real | | Outcome and date | The result | | Reason (if lost) | Price, schedule, relationship, capability, no response | | Winning price, if known | The single most useful competitive data point |

Weighting by dollars matters. Winning nine $2,000 jobs and losing one $400,000 contract is a 90 percent win rate and a bad quarter.

What the segments tell you

The aggregate number is nearly useless on its own. The splits are where the information is.

Public versus private. Public work is usually low-bid and published; private work is negotiated and relationship-driven. A company that wins 45 percent of private bids and 10 percent of public ones is not bad at bidding — it is priced for service, and it should stop spending estimating hours on hard-bid work it does not want. Bidding municipal traffic control contracts covers that lane specifically.

By prime. Win rates against specific general contractors vary enormously and are often about being on the invitation list rather than about price. A prime you win 5 percent of work from may be using you to cover its bid.

By job type. Losing all your flagging-only bids while winning device-heavy jobs usually means your labour rate is high relative to competitors and your device rates are not. That is a concrete pricing conversation. Traffic control device rental rates and pricing and flagger labour cost per hour: the loaded rate cover both sides.

By response time. Time from request to quote sent, plotted against outcome, frequently shows the sharpest relationship of anything in this data — particularly for short-notice work. If the pattern shows up, quoting traffic control jobs faster is a higher-return project than a pricing review.

By season. Win rate typically rises in the shoulder months when everyone has capacity, and falls at peak when bidders are selective. Managing seasonality in a traffic control business covers planning around it.

The trap: chasing the number

Win rate is a metric that can be improved by doing something stupid. Cut prices and it goes up. Bid only work you are certain of and it goes up. Neither makes you money.

Read it alongside two other numbers:

Margin on won work. Rising win rate with falling margin is the signature of buying work. Job costing for traffic control companies is the discipline that lets you see it.

Estimating hours per bid. If preparing a bid costs four hours and you win one in twelve, that is 48 hours of estimating per win. Sometimes the right answer is to bid less and better. Bid/no-bid decisions for traffic control companies covers the filter.

Where public bid tabulations are published — state DOTs commonly publish letting results, as TxDOT does through its contract letting pages, and federal opportunities and awards are posted on SAM.gov — you can see exactly how close you were. DOT prequalification for traffic control contractors covers getting on the list in the first place. Losing by 2 percent repeatedly is a pricing adjustment. Losing by 40 percent means you are bidding a different job than everyone else, which is usually a scope misunderstanding rather than a price problem — and worth understanding before you cut anything.

Reasons are worth more than rates

The lost-reason field earns its keep. Four codes are usually enough: price, schedule/capacity, capability/prequalification, no response.

Each points somewhere different:

• Price — a pricing or cost conversation.
• Schedule — you were busy, which may be fine, or a capacity planning signal.
• Capability — prequalification, certifications, equipment. Safety records and EMR in prequalification covers a common blocker.
• No response — relationship or follow-up, and the cheapest to fix.

Where the data should live

Ideally in the same place quotes are produced, so recording an outcome is one click rather than a separate spreadsheet somebody maintains until they get busy. The practical test is whether the person who sends quotes can mark an outcome in under ten seconds. If not, the data will be incomplete within two months, and incomplete win-rate data is worse than none — it is confidently wrong.

Traffic OS keeps quotes with the customer and job record so outcomes are recorded where the quote was made, and rate sheets sit behind the quote so the same job priced twice comes out the same. Flat monthly tiers — $499, $949 and $1,499 as of September 2026, no per-user charge — are on the pricing page; the features page covers quoting and job costing, and a walkthrough can use your own bid history.

A quarterly routine

1. Count bids sent, won, lost, no decision — by count and by dollars.
2. Split by customer type, job type and estimator.
3. Look at response time against outcome.
4. Read the lost reasons; pick the largest one.
5. Compare margin on won work against the same quarter last year.
6. Decide one thing to change — a rate, a segment to stop bidding, or a response-time target — and write it down so next quarter can test it.

The number itself is not the point. The point is that a company that measures this stops guessing about why work is or is not coming in, and starts making one deliberate change a quarter instead of reacting to whichever job it lost most recently.