Most bid teams do not have a pipeline. They have a folder of PDFs, a deadline spreadsheet, and a standing argument about which tender everyone is supposed to be working on. The symptom is always the same: the team is busy every week and surprised every quarter.
A bid pipeline fixes this the way a sales pipeline fixed field sales: by making the funnel explicit. Every opportunity sits in exactly one stage, moves forward only by meeting criteria, and exits with a recorded outcome. Once that structure exists, the numbers it produces (conversion rates, win rate, coverage) start answering the questions the folder of PDFs never could.
The three qualification gates
Every bid you prepare costs real money before a single mark is scored: estimating hours, document fees, and a bond. So the first job of a pipeline is not to help you bid more; it is to help you bid less, on better tenders. Three gates do most of the filtering:
- Fit. Is this scope actually what you deliver, or adjacent to it? Does the buyer have a history of awarding to firms your size, or do its bid tables show the same three large incumbents every time? Do you meet the eligibility conditions as written: classifications, certifications, local-content thresholds?
- Capacity. If you won it, could you deliver it alongside the current book? And can you afford to pursue it: a serious proposal costs something like 60 hours of estimating and writing, and those hours come out of the same team delivering current contracts.
- Bond affordability. A 5 percent bid bond on a 3M tender parks 150k of cash or credit line for months, win or lose, and a win typically converts it into a larger performance bond. A tender whose bond you cannot carry is not an opportunity, whatever the margin looks like.
A tender that passes all three gates is qualified. One that fails any gate exits now, cheaply, instead of at the deadline, expensively.
Bid or no bid, as a discipline
Gates filter what you could bid. The bid or no bid decision picks what you will bid, and it only works as a discipline: a short scheduled meeting, a simple score, and a recorded reason either way.
Score each qualified tender on three axes: fit (how squarely it matches your delivery strengths), winnability (bidder density, incumbency, and your relationship with the buyer, the material from the earlier lessons), and economics (expected margin at the price the benchmarks lesson says would win). Set a threshold and hold it. The score matters less than the habit: a written reason for every no bid, revisited quarterly, is how a team learns what its wins have in common.
The hardest part is cultural. Declining a biddable tender feels like losing, so undisciplined teams bid everything and dilute their estimating hours across proposals they were never going to win. Killing a weak bid early is the highest-return decision a bid team makes, and a pipeline makes the kill visible and reversible instead of silent.
Stage design: six stages, each with an exit rule
Stages only mean something if moving forward requires evidence. A working design for a bid pipeline:
- Identified. The tender exists, the deadline and bond are known, a source link is attached. That is all this stage claims.
- Qualified. It passed the three gates and survived bid or no bid. An owner is assigned.
- Bid prepared. Costed, priced against the category benchmarks, documents drafted, bond arranged. This is where most of the labor lives.
- Submitted. Delivered before the deadline, receipt confirmed. Nothing left to do but wait and clarify.
- Won / Lost. Terminal outcomes, each with a reason recorded. Where the portal publishes the bid table, record the winning price and your distance from it; that is next year's benchmark data.
- No bid. Also a terminal outcome, never a deletion. A no bid with a recorded reason is data; a deleted tender is amnesia.
This is the point where a spreadsheet starts to strain: six stages, owners, deadlines, bond amounts, and outcome reasons across dozens of live tenders. In Ishara, the tenders screen has a multi-select tray that pushes chosen tenders straight into a bid-team pipeline, where stage-entry requirements enforce the exit rules above and every won or lost outcome is recorded on the deal. One screen, same funnel:
Win-rate math and pipeline coverage
Once outcomes accumulate, the pipeline produces the number that sizes everything else: your win rate. And the win rate turns any revenue target into a top-of-funnel requirement. If you win 1 submission in 4, a revenue target needs 4x that value in submissions. That multiple is your pipeline coverage, and running below it is how quarters get quietly lost months in advance.
Worked example: sizing the funnel for a 12M year
Your target is QAR 12M in new contract value next year. The average award in your categories runs about 1.5M, so the target is 8 wins. All figures are illustrative round numbers.
Your pipeline history gives three conversion rates: identified to qualified 40 percent (the gates are doing their job), qualified to submitted 50 percent (bid or no bid kills half), and submitted to won 25 percent (you win 1 in 4).
Chain them backwards: 8 wins ÷ 0.25 = 32 submissions. 32 ÷ 0.50 = 64 qualified. 64 ÷ 0.40 = 160 identified. The funnel must swallow about 13 identified tenders a month to hit the target.
Now the checks. Coverage: 32 submissions × 1.5M = 48M of submitted value against a 12M target, exactly the 4x your 25 percent win rate demands. Capacity: 32 proposals at roughly 60 hours each is about 1,900 hours, most of one full-time estimator, which loops straight back to the capacity gate. If you cannot staff 32 bids, the honest levers are a higher win rate through better selection, or a smaller target. More bids with thinner effort lowers the win rate and shrinks the funnel from the inside.
Two habits keep the math honest. Measure conversion rates from your own recorded outcomes, not from this lesson's illustration; a team that never logs its no bids will flatter its qualified-to-submitted rate and starve the top of the funnel. And re-run the arithmetic quarterly, because win rates move with category mix. The bid pipeline calculator runs this chain for any target, rates, and average award size.
The funnel does not end at Won
The day you win a contract, two clocks start. The first is delivery. The second is the recompete: your new contract has an end date, and that end date is a future tender with you as the incumbent.
Feed it back into the pipeline deliberately. Create the recompete as an identified opportunity the week you win, dated 6 to 12 months before expiry, the window the recompete timing lesson mapped. Between now and then, incumbency defense is mostly delivery made visible: documented performance, renewals of certifications the eligibility rules will ask for again, and attention to the buyer's other awards, because shifts in its supplier mix show up in award data before they show up in meetings.
And remember the symmetry: every contract you won is sitting on a competitor's radar with the same expiry date. The difference between an incumbent that holds and one that gets displaced is usually which side treated the recompete as a pipeline entry from day one.
That closes the course. You know how these procurement systems work, how to read a bid table, when expiring contracts create openings, how to price against history, and now how to run the whole flow as a funnel with math attached. The pipeline is the piece that compounds: every recorded outcome makes the next gate call, price, and coverage estimate a little less like guessing.