A slow quote rarely loses on price. It loses because it arrives after the buyer has already decided, so it is never compared at all. The cost does not show up anywhere: the order you lost to speed looks exactly like an enquiry that was never serious. This post sets out why speed matters, how to put a number on it with your own data, and what to change first. Where there is arithmetic, the numbers are examples, labelled as such.
How buyers actually behave
An engineer who needs a part has a deadline, a file and usually more than one shop to ask. They send the file out, and they keep working on the rest of the project. When a credible price arrives that fits the budget and the date, many will order and move on, because the part is one task among dozens and the next quote is unlikely to be dramatically cheaper.
That is the whole mechanism. A buyer does not need to be impatient for speed to matter. They only need to have something else to do.
Two things have made it stronger. Marketplaces answer in seconds, so the first credible price is often one the buyer did not have to wait for at all. And buyers increasingly prefer to get there without talking to anyone: in a 2025 Gartner survey of B2B buyers, 61 % said they prefer an overall rep-free buying experience (Gartner).
How slow are businesses, generally?
There is little published research on response times in manufacturing specifically, so treat the nearest evidence with care. A 2011 Harvard Business Review study of online sales leads in general (not manufacturing) sent a web enquiry to 2,241 US companies: 37 % replied within an hour, and among those that replied within 30 days the average response took 42 hours (Harvard Business Review; figures as reported by SmartCompany). Different industry, different decade, but the point carries: many businesses take far longer to reply than they think.
Where the hours go in a manual quote
The pricing itself is rarely slow. The waiting around it is.
| Step | Typical delay | Why |
|---|---|---|
| Request lands in a shared inbox | Minutes to hours | Nobody owns it until someone notices |
| Waits for the estimator | Hours to a day | The estimator is also running the floor |
| File opened, checked, sliced or measured | Minutes to an hour | Broken files, wrong units, missing drawings |
| Questions back to the buyer | A day per round | Material, quantity or tolerance not stated |
| Price calculated and written up | Minutes | The quick part |
| Request arrives late on a Friday | Over the weekend | The calendar |
None of these is anyone’s fault. Together they turn ten minutes of pricing into a day or two of elapsed time.
Putting a number on it
You can estimate what slowness costs from your own records. You need four things for each quote: when the request arrived, when the price went back, whether it was won, and its value.
Group the quotes by response time and compare win rates. Then multiply the gap by your volume and average order. The example below shows the arithmetic; every figure in it is assumed, not measured.
Worked example: a shop answering 60 requests a month (all figures assumed for illustration)
| Line | Calculation | Amount |
|---|---|---|
| Requests a month | Assumed | 60 |
| Average order value | Assumed | €500 |
| Win rate when answered next day | Assumed | 20 % |
| Win rate when answered the same hour | Assumed | 30 % |
| Orders a month, next day | 60 × 20 % | 12 |
| Orders a month, same hour | 60 × 30 % | 18 |
| Extra revenue a month | 6 × €500 | €3,000 |
| Extra contribution a year at a 30 % margin | €3,000 × 12 × 30 % | €10,800 |
The gap between 20 % and 30 % is an assumption. Your records will give you your own figure, and it may be larger or smaller. The point of the exercise is that a modest change in win rate, multiplied by a year of requests, is usually worth far more than the subscription to anything that would produce it.
There is a second cost on the other side of the ledger: the estimator’s time. At 30 minutes a request, 60 requests take 30 hours a month of the person who most understands the work. Some of that time is judgement worth paying for. Much of it is opening files.
What to change first
You do not have to answer everything instantly. You have to stop routine parts from waiting.
- Measure it. Add two timestamps to every quote: request in, price out. Without them you are guessing.
- Answer routine parts at once. Most shops have a large set of parts that are routine to price: the materials you stock, the sizes you know, quantities you run every week. An instant quote on your own website prices those from your own rates while the buyer is still on the page. Quotometry does this for FDM, resin, SLS and MJF.
- Acknowledge the rest within the hour. A short reply with a date for the price (“you will have it by Thursday noon”) keeps the buyer from deciding without you. Then keep the date.
- Ask your questions in one go. Every round of questions costs a day. A request form that asks for material, quantity, finish and deadline up front removes most of them.
- Own the inbox. One named person, or a rota, so a request never waits for someone to notice it.
- Watch the result. Compare win rates before and after. If the gap you measured closes, you have your answer.
What speed does not fix
Speed gets your quote compared. It does not win the comparison on its own. A quick quote that is unclear, priced from guesswork or missing the buyer’s deadline still loses. The fast answer has to be a good answer: the part shown, the assumptions stated, the lead time real, any manufacturability problems flagged. The hidden cost of quoting by e-mail covers the costs that a slow, manual process hides, and marketplaces or your own website covers what happens when a buyer finds that a marketplace answers faster than you do.
For the measurements in more depth (win rate by response time, process, quantity and customer) see quote win rate.
Questions people ask
How quickly should a manufacturing shop answer an RFQ?
As quickly as the buyer's alternatives do. For routine parts that now means seconds, because marketplaces and instant quote widgets answer while the buyer is still on the page. For complex parts, acknowledge the request within the hour with a date for the price, and keep that date. Measure your own response times and win rates to see where your threshold is.
How do I know if slow quoting is losing me orders?
Record, for every quote, when the request arrived and when the price went back, and whether it became an order. After a few months, compare the win rate of quotes answered the same day with those answered later. If the slower quotes win noticeably less often, the gap, multiplied by your average order value, is what slowness costs.
Quotometry prices every part from your own machines and rates, on your website, in seconds.