We looked at 1,125 quotes from 43 small businesses. Of the quotes that had at least six months to be answered, 45.5% never got an answer on the record, and almost none were ever turned down in writing. This is a study of what invoicing software knows about a quote, which turns out to be very little.
Key figures
1,125 quotes from 43 small businesses, last 24 months. Free to reuse under CC-BY with attribution. Download the CSV
Every quote sent at least six months ago, so each one has had a fair chance to be answered. Without that cut the unanswered share is flattered by quotes too recent for anyone to have replied to yet.
| Outcome | Quotes | Businesses | Share |
|---|---|---|---|
| Answered yes, and became a job | 475 | 29 | 54.3% |
| Never answered either way | 398 | 30 | 45.5% |
| Formally declined | Suppressed. Too few businesses have ever marked a quote declined for the bucket to clear our 25-business threshold. | ||
Count each business once and the number goes up, not down. The 45.5% above pools every quote together, which lets a business that sends a lot of them pull the figure around. Counting each business once instead, the median business never hears back on 50.0% of the quotes it sends. We report both, and the business-level one is the fairer description of a typical small business.
That last row is the finding, and it needs its mechanism stated or it is worthless. Declining is something the client does on the online copy of the quote. The owner has no decline button at all, and as the next section shows, about two thirds of clients never open that online copy. So the near-zero count does not establish that clients rarely refuse. It establishes that a refusal almost never reaches the record, and that an owner watching the software for a no will be watching an empty space whatever the client decides.
From our quote data
Half of unanswered quotes have been waiting more than 383 days, and one in ten more than 594 (521 quotes, 35 businesses). Silence is not a decision still being made.
Source: Billbooks quote study, 1,125 quotes from 43 small businesses, last 24 months (aggregate only, figures shown only where 25+ businesses and 100+ documents contribute). See the study
The obvious reading of an unanswered quote is that the client has not got round to it yet. Half of the quotes still sitting unanswered were sent more than 383 days ago, and one in ten more than 594. Read that figure carefully, because it is partly a property of the window. Ages are measured to the day the study ran, across a 24-month window, so if quotes went out at a steady rate and none were ever answered you would mechanically see a median near a year anyway. What it does establish is that these are not fresh quotes awaiting a decision: the record has been sitting untouched for a long time, and nothing in the software is going to close it.
From our quote data
About 64% of quotes are never opened through the online link at all (725 of 1,125 quotes, 35 businesses). That is not proof they were unread, since most are also sent as a PDF by email.
Source: Billbooks quote study, 1,125 quotes from 43 small businesses, last 24 months (aggregate only, figures shown only where 25+ businesses and 100+ documents contribute). See the study
A word on what "opened" records here. The flag is set when the document's own online link is loaded, and nothing checks who loaded it. It therefore undercounts one way and overcounts the other: a client who reads the PDF attached to the email and never clicks through counts as never opened, while an owner previewing their own link, or a mail client prefetching the URL, counts as opened. Read it as a record of whether the online copy was fetched, not as proof of whether a human read the document.
So this is not evidence that two thirds of quotes go unread. It is evidence that for two thirds of quotes, the sender has no idea either way.
From our quote data
Nearly three quarters of quotes carry no expiry date (836 of 1,125 quotes, 34 businesses), so last year's price is still technically on the table.
Source: Billbooks quote study, 1,125 quotes from 43 small businesses, last 24 months (aggregate only, figures shown only where 25+ businesses and 100+ documents contribute). See the study
A quote without an expiry date does not stop being a quote. Combined with the ages above, that means a large number of small businesses have live offers outstanding at prices they set more than a year ago, against costs that have moved since.
Stated because a study that only reports what worked is not worth trusting. Each of these was computed, failed our own threshold of 25 businesses and 100 documents, and was therefore dropped.
Whether opening a quote predicts an answer. The obvious next question, and the cross-tab splits the population into four cells of 23, 23, 27 and 29 businesses. Two of the four clear our threshold and two do not, and a comparison is only as publishable as its weakest cell. We can say how many quotes were opened. We cannot say whether opening one changes the outcome.
Whether an expiry date changes the outcome. Same problem, and worse: the expiry-set and answered cell holds 110 documents from 19 businesses, six businesses short.
How long a won quote takes to become an invoice. Held back by judgement rather than by the rule. 546 converted quotes from 27 businesses does clear our threshold, but by two businesses. A figure sitting that close to the bar can fall back under it as the 24-month window moves forward, and we would rather publish it once it is comfortably clear than publish it now and withdraw it later.
Whether chasing a quote works. Follow-ups are switched on for 36 quotes across 8 businesses. Far too few, and self-selected in the same way payment reminders are: an owner turns them on for the clients they already expect to chase, so the setting marks an expectation rather than creating an outcome. Our payment study refused to publish the equivalent reminder comparison for exactly this reason, and refusing it here too is the consistent answer rather than a convenient one.
Why any individual quote went unanswered. The data records the absence of a reply. It cannot record a reason, and estimating one from another field would be inventing the dimension rather than measuring it.
Source. Billbooks quote records, last 24 months. Population.
Quotes not left as a draft, with samples and demonstration data excluded. There is no "sent" flag or
send date in the record, so "sent" throughout this study means a quote that left draft, and "sent at
least six months ago" means one dated at least six months ago.
Outcome is inferred from the document, not from a recorded client decision. A quote
that became an invoice, or was marked approved, counts as a yes; one still sitting as sent counts as
never answered. That inference is this study's main limitation and it cuts both ways:
an owner who was told yes by phone and never updated the quote shows here as unanswered, and so does
an owner who was told no by phone. The 398 unanswered quotes therefore contain phone yeses, phone
noes, text messages and genuine silence, and this data cannot separate them. Every outcome here
except conversion to an invoice depends on the client using the online copy of the quote, which most
do not, so this study measures what the software recorded rather than what clients did.
Fair-chance cohort. The headline outcome uses only quotes sent at least six months
ago. Privacy. Aggregate only. Every reported bucket holds at least 25 distinct
businesses and at least 100 documents; smaller buckets are suppressed rather than rounded, and the
suppressed ones are listed above. No business, client, document or free-text field is read. The
query is committed at aeo/quote-study.sql and runs against a restored copy, never
against production. These figures were computed on 25 September 2026 against a restore of
the 24 September 2026 backup. Every window here is relative to the run date, so a later
run will move them.
The quote count is larger than the effective sample. 1,125 quotes come from 43 businesses, and they are not evenly spread: the largest single contributor accounts for 223 of them, just under a fifth of everything here. That is why the business-level rate is published beside the pooled one above. These are businesses that use invoicing software and choose to send quotes through it, which is not the same as all small businesses. The base is 43 businesses, far smaller than the 180 behind our invoice payment study, and it should carry correspondingly less weight. Ages of unanswered quotes are measured inside a 24-month window, so the distribution is truncated at the top: the true typical wait is at least 383 days and may be longer. Billbooks does not record an industry for a business, so none of this can be broken down by trade.
Three things follow from this data, and none of them needs our software.
Stop reading silence as a no. It is the most common outcome by far, and the record cannot tell you what it means. Treating it as a rejection writes off a large share of a pipeline on a signal that carries almost nothing.
Put an expiry date on the quote. Three quarters carry none, which is what leaves a year-old price technically still on the table. A date gives both sides a reason to close the file.
Decide in advance when you will follow up, and write it on the quote. A quote with no next step on it is an invitation to think about it later, and half of the unanswered ones have been sitting for more than a year. If you want the arithmetic done for you, our quote follow-up planner is free and needs no signup.
Source: Billbooks quote study, 1,125 quotes from 43 small businesses, last 24 months (aggregate only, figures shown only where 25+ businesses and 100+ documents contribute). Free to cite and reuse under CC-BY 4.0 with attribution to Billbooks. The aggregate table is published as a CSV.
Most are never answered. Across 875 quotes sent at least six months ago by 35 small businesses, 45.5% got no answer either way, and counting each business once the median business never hears back on half the quotes it sends. A formal decline is so rare it falls below our own reporting threshold.
Declining is an action the client takes on the online copy of the quote, and about two thirds of clients never open that copy. The owner has no decline button at all. So a refusal almost never reaches the record, which is not the same as clients rarely refusing.
The record cannot tell you. Quotes that went unanswered contain agreements made by phone, refusals made by phone, and genuine silence, and invoicing software separates none of them. What it does show is that half of unanswered quotes have sat untouched for more than 383 days, so nothing is going to resolve them on its own.
Of quotes sent at least six months ago, 45.5% were never answered either way: 398 of 875 quotes across 30 small businesses. Counting each business once rather than pooling, the median business never hears back on half the quotes it sends.
Almost never in writing, but the mechanism matters more than the number. Declining is an action the client takes on the online copy of the quote, and about two thirds of clients never open it. So a refusal rarely reaches the record, which is not the same as clients rarely refusing.
Half of unanswered quotes have been waiting more than 383 days, and one in ten more than 594 days, across 521 quotes from 35 businesses. Silence is not a decision still being made.
This study cannot say. Only 26% of quotes carry an expiry date, and splitting the outcome by whether one was set produces four groups of which two fall below our 25-business reporting threshold. A comparison is only as publishable as its weakest group, so it was suppressed rather than published.