Quote Turnaround Time: Where the Days Actually Go

The owner asks why the quote took three weeks. The estimator, if honest, does not answer with a list of work. He answers with a list of waits: four days before anyone opened the package, a question to the customer that went out in the second week and came back as an addendum in the third, two supplier quotes that arrived the day before the deadline, and an afternoon in the general manager's inbox waiting for a signature. The cost build itself, the part everyone calls estimating, took a day and a half. Quote turnaround time, measured as one number from request in to quote out, hides all of that, and the number it produces for an engineered job is usually weeks.

This piece takes the number apart. It uses public bid packages, because the customer's clock is written into them and almost nobody reads that part first, and it uses what we have measured on the reading step, because that is the block nobody sees. I build software that does the reading, so take the last section with that in mind. Most of what shortens a quote is not software.

In brief

• Quote turnaround time is three clocks, not one: the customer's window, your internal clock, and the validity clock that starts the day you submit. Most companies measure only the second and are surprised by the first and third.
• Inside a multi week engineered quote, the cost build is a day or two. The rest is reading, waiting on the customer, waiting on suppliers, and queueing for review. The waits are sequential only because the reading happens late.
• The customer's document sets deadlines the seller discovers too late: 21 days to bid with questions cut off five days before opening; a package that stops answering questions seven days out; a price that must hold 60 days for equipment due 60 weeks later.
• The useful metric is time to first number: how long from arrival until somebody can state a rough cost with a written list of assumptions. It exposes the reading, which the total hides.
• A page cited requirements list on day one compresses everything downstream, because questions and supplier requests can leave the same day and review can review the reading instead of the total.

What quote turnaround time actually measures

The definition is simple: the elapsed time between a customer's request arriving and the quote leaving. The glossaries stop there, and that is why they are not useful to a custom manufacturer. On an engineered quote the number is the sum of three different clocks, and only one of them is yours.

The first is the customer's window. A bid package comes with a due date, and often with a second, earlier date after which questions will not be answered. Your turnaround time cannot exceed the window, whatever your process, and if your questions go out after the cutoff your process has already failed on that job. The second is your internal clock: the days between the package arriving and the quote being signed, which is the number owners track and the number this article decomposes. The third starts the day you submit and runs until the customer decides, and it is the clock the validity clause governs. On one public case, a $252,000 equipment quote was asked to hold its price for 180 calendar days and was still being held to it on day 260. A company that measures turnaround as one number will never see that clock, and it is often the most expensive of the three.

So the honest measurement has three stamps: the date the package arrived, the date the quote left, and the date the customer decided. Everything in this article is about the middle interval, but the other two frame it, and the first one is set by somebody else.

Quote lead time, delivery lead time and turnaround: three phrases, one confusion

A good share of the people searching for quote lead time are buyers trying to work out what the lead time printed on a quote means, and the answer is that it is not about the quote at all. Lead time on a quote is the delivery promise: how long after the order the goods arrive. On the switchgear package above it was sixty weeks. Quote turnaround time is the time it took to produce the quote. Quotation lead time, in the procurement systems that use the phrase, usually means the same as turnaround, the days between request and offer. Three phrases, two of which describe your clock and one of which describes the factory's.

The confusion matters to sellers for one reason. A customer who reads a long delivery lead time and a slow quote turnaround on the same document draws one conclusion about the company, and it is not a flattering one. The delivery lead time is usually honest and hard to move. The turnaround is the part you control, and it is the part they see first.

Where the days go on an engineered quote

Here is a composite of the engineered quotes we have watched, laid against the 21 day window a fire district gave bidders for a 99 page pumper truck request: invitation dated January 12, bids due February 2, questions required at least five days before the opening. The days are illustrative and the pattern is not. The second column is the sequence as it usually happens. The third is the same work with the reading moved to the front.

Two things stand out. The first is that almost nothing in the left column is slow because the work is slow. It is slow because each segment waits for the one before it, and the one before it is waiting for the reading, which nobody scheduled. The second is that the right column is not faster because anyone worked harder. It is faster because the reading produced an output on day one, and every other segment could start from that output instead of from the package.

RFQ turnaround time from the buyer's side

The customer has a clock too, and it is written into the package, usually in the front end documents that sellers read last. Reading it first changes what turnaround time means on that job.

The fire district's 21 days came with a question cutoff five days before opening, which means the real window for finding a contradiction and asking about it was 16 days, and a seller whose reading reached the warranty pages in week three had no route to an answer. The elevated tank package says it more sharply: questions received less than seven days before bid opening may not be answered, and only responses issued as an addendum are binding, so an answer by phone is not an answer at all. A federal crane solicitation offers exactly one professional route for the sentence that could double the machine, a written question to the contracting officer before bids are due. On every one of these, the buyer's clock rewards the seller who read early and punishes the one who read thoroughly but late.

The buyer's clock also runs after submission. On a switchgear package from a municipal light department, bid prices had to stay firm for sixty days from opening, while the equipment was due sixty weeks after an order that would not be placed on opening day. The seller's turnaround time ended at submission. The seller's exposure did not. RFQ turnaround time, the phrase procurement teams use for the same interval from their side, includes their evaluation, their board meeting and their funding cycle, and a seller who has not priced that clock has priced only two of the three.

The practical move is to build the customer's clock into the register on day one: due date, question cutoff, addendum rule, validity term, delivery term. Four lines, all from the front end documents, and they decide the schedule of everything else.

Time to first number

The number that owners track, request in to quote out, is the right number to report and the wrong number to manage by, because it cannot tell you where the days went. The number to manage by is time to first number: the elapsed time from the package arriving until somebody in the company can say, with a written list of assumptions, roughly what this job will cost.

Time to first number does two things the total cannot. It measures the reading, because a first number with written assumptions is only possible once the requirements have been found, so the interval is mostly reading and triage. And it separates your clock from the customer's, because the first number does not wait on addenda or supplier quotes; it brackets them. A company whose first number arrives on day 2 and whose quote leaves on day 20 has a waiting problem. A company whose first number arrives on day 12 has a reading problem, and no amount of supplier chasing will fix it. And a company that cannot produce a first number at all until the suppliers answer has a third problem, which is that its own rules do not exist in a form anyone can bracket with, and that one is the most expensive of the three.

Measuring it needs three timestamps and no software: when the package arrived, when the first bracketed number with assumptions was written down, and when the quote left. Keep them on the quote itself, for every quote, for a quarter. The pattern that emerges is the first honest picture most companies get of their own turnaround, and in every company we have looked at, the first interval is the one that surprises the owner.

The reading step with an output

The reason the right hand column of the table works is a single change: the reading produces a document. Not notes in a margin, not highlights, not the estimator's memory of what page 79 said. A list, one row per requirement that changes the price, each with its page number and each with a flag for the ones that need a question, a supplier, or a decision.

Once that list exists on day one or two, three things happen that cannot happen otherwise. The questions to the customer leave immediately, because the contradictions are on the list, and they leave before the cutoff. The supplier requests leave the same day, because the bought out items are on the list with their specifications attached, and suppliers get two weeks instead of three days. And the review changes from twenty minutes on a total to twenty minutes on the list, which is where the nuclear standard and the phantom foam system would have been caught, and which is the only review that can catch them.

The list also fixes the problem the label maker wrote to us about, the one where the whole process lives in one person's head. The estimator who has read thirty of these packages still reads them faster than anyone else. But the output is now a document someone else can check, extend and cost against, and the day the estimator retires the company keeps the lists.

The same clock with twenty five estimators and four plants

At a single plant with two estimators, the queue is visible. Everyone knows which packages are open on whose desk. At a group with four plants and twenty five people quoting, turnaround time is dominated by a queue nobody can see, and the segment table above gains a row at the top: which plant, which estimator, and how long the package spent being forwarded before it landed.

Three things happen at scale that do not happen at a single desk. Packages arrive at sales, at a plant, at a general inbox and at a personal one, and the first day is spent working out who owns it; a package addressed to the wrong plant can lose a week before anyone opens it. The reading is done four different ways by four lineages of estimator, so the same specification produces four different requirements lists, and the supplier requests go out on four different days. And the review queue is not one person's afternoon but a chain: plant estimator, plant manager, group commercial lead, each adding a day and each reviewing the total because the reading is not visible to be reviewed.

The fixes are the same as for the small company with two additions. One intake point, with a same day rule for opening and registering the package, whichever inbox it landed in. And one written pricing identity across the plants, so that a group review is a comparison against a rule rather than a re-estimate. Groups that run the nine station audit across plants usually find that their turnaround number is really four numbers, and that the slowest plant is not the one with the hardest packages but the one whose estimator reads last.

How to improve quote response time

In order of payoff, from what we have watched work. None of these is software, and the first three cost nothing.

1. Read the front end documents on the day the package arrives. Due date, question cutoff, addendum rule, validity, delivery, damages. Ten minutes, and it sets the schedule for everything else. Most sellers find the question cutoff after it has passed.
2. Make bid or no bid a same day decision from the document, not the cover letter. A package that will be declined on day 9 has cost nine days of a queue that other quotes were waiting in. The pre bid register is the tool.
3. Give the reading an owner, a deadline and an output. One person, the first two days, a requirements list with page numbers. This is the whole change. Everything below depends on it.
4. Send questions and supplier requests from the list, the same day it exists. Both are waits you cannot shorten once they start, so start them first. Bracket the answers in the cost build rather than waiting for them.
5. Log assumptions instead of waiting for answers. Where the document is silent and the cutoff has passed, write the assumption on the quote and price to it. A quote with a clear written assumption beats a late quote with a phone call behind it, and the customer can see exactly what they are buying.
6. Write the pricing rules down so the review queue disappears. The senior person's afternoon is spent checking whether the margin logic was applied, because it lives in their head. If it lives on paper, the check is a comparison and takes minutes, and the queue goes away. The tribal knowledge piece covers how to get the rules out.
7. Stamp three dates on every quote and look at them quarterly. Arrival, first number, submission. The distribution tells you which of the six moves above you still have not made.

Three things do not work, and all three are common. Pressure on the estimator shortens the reading, which is the one segment that should not be shortened, and it produces the foam system nobody priced. A faster spreadsheet shortens the cost build, which was already a day or two; it cannot touch the waits. And a configurator for engineered work, bought to make quoting faster, moves the reading into an engineer's inbox and makes it invisible, which is slower, as the CPQ vs ERP piece explains at length.

What to tell the customer while they wait

Turnaround is also something the customer experiences, and the experience can be managed independently of the number. The seller who acknowledges the package on the day it arrives, names the person reading it, and states when a number will follow has already done something most competitors will not, and it costs an email.

The first number is the second thing to send, on the jobs where the customer will take it. Many buyers of engineered equipment, especially the consulting engineers who write the packages, are happy with a budgetary figure and a written list of assumptions ahead of the formal bid, because it tells them early whether their own estimate is in range. A first number on day three with six assumptions attached is worth more to that reader than a firm price on day twenty, and it is the same document your own cost build is about to start from. Send it, marked as what it is.

The third thing to send, when it is true, is a no. A same day decline on a package you were never going to win saves the customer a wasted evaluation and saves you nine days of queue, and it is remembered. Companies that decline early and clearly get invited back. Companies that go quiet for three weeks and then decline do not, and the difference shows up in the win rate two years later where nobody will connect it to turnaround.

Quoting KPIs worth tracking

Six, with definitions, and no benchmarks, because the public benchmarks that circulate come from procurement glossaries and freight forwarders and none of them measured an engineered quote. Your own numbers, over a quarter, are the only ones that mean anything, and they are enough.

Win rate is deliberately not on the list, not because it does not matter but because it is a lagging, noisy number that depends on price, relationship and competition as much as on speed. Track it by lane, and read it beside the six above rather than instead of them. What we can say without data is narrow and certain: the quote that arrives after the deadline loses, and the quote that arrives fast with the foam system unpriced wins and loses money. Speed is worth having only with the reading intact.

Volume changes the clock

Everything above is about the engineered quote: one document, hundreds of pages, a product that does not exist yet, weeks of window. A different kind of company writes to us with a different clock. The managing director of a custom label maker in central Europe quotes small, numerous, technically known jobs, and his turnaround problem is measured in hours, not weeks, because his customers expect a price the same day and his one experienced estimator is the queue.

The two problems share a name and almost nothing else. His is a routing and cost model problem, best solved by the print industry's own management systems, which price a job across known machines in minutes once the rates are in them. Ours is a reading problem, and no routing engine reads a 99 page specification. A company with both kinds of quote under one roof, and many have, should measure them separately and fix them separately, because a single turnaround number across both lanes will be dominated by the engineered jobs and will hide the fact that the quick ones are late too. The buy, build or combine piece sets out which system belongs in which lane.

Where software helps, and where it does not

The block that software changes is the reading, and only the reading. On a 179 page marine package we read publicly, the engine found and cited 214 requirements in 44 seconds and flagged seven internal conflicts. On a 668 page municipal wastewater package it found the 14 sections that carried requirements for the equipment in 102 seconds. Both numbers are on the pilot page with the packages named, and both are the first day of the left hand column above, compressed to the time it takes to find a pen.

What that buys is not a faster quote by itself. It is the requirements list on the morning the package arrives, which is what lets the questions leave, the supplier requests leave, and the cost build start with brackets instead of waiting. Every downstream wait is still a wait. The customer still answers by addendum on their schedule. The supplier still takes a week. The review still needs a person, and the assumptions still need somebody willing to sign them. Software moves the reading to day one. The rest of the right hand column is process, and a company that has the process can get most of the way there with a highlighter and a rule about who reads first.

What we do, and where the audit fits

Mavlon builds reading engines for custom manufacturers, and turnaround is the symptom that usually starts the conversation. It is rarely the problem. The problem is a reading step with no owner and no output, and a company can fix that before it buys anything. If you want to see where your own days go, stamp the three dates on the next ten quotes. If you want someone to run the instruments with you, the quoting system audit starts with your archive and your estimators and ends in one of three honest answers, one of which is that you do not need us. Bring the package you least want to read again, and the calendar that shows when it arrived.