Quick answer: There is no credible current benchmark for how fast freight forwarders respond to RFQs. The most-cited figure — 90 hours — comes from a 2015 study that tested 16 forwarders on LCL shipments from China. Nothing comparable has replaced it, and no public research links response time to win rate.
Ask any forwarder whether quoting speed matters and you will get an immediate yes. Ask what "fast" means on their trade lane, and the conversation stops.
I went looking for the answer — a defensible number for how long a freight forwarder takes to respond to an RFQ, and what each hour of delay costs in win rate. What I found instead is worth writing down, because the absence is more interesting than any number would have been.
Search for freight quote response times and you will land, directly or indirectly, on one figure: forwarders take an average of 90 hours to return a quote.
It is a real finding. It is also from June 2015.
FreightWaves reported the research on 11 June 2015. Freightos — a freight booking software company — submitted online quote requests for less-than-containerload shipments from China to 16 global forwarders and recorded what happened.
| Measure | Result |
|---|---|
| Average time to receive a quote | 90 hours |
| Fastest confirmation of receipt | 7 hours |
| Fastest actual quote | 30 hours |
| Slowest actual quote | 840 hours |
| Forwarders that ultimately quoted | 45% |
| Forwarders that never responded | 3 of 16 |
| Did not automatically confirm receipt | 75% |
| Followed up after quoting | 5 companies |
Read the right-hand column and the average stops being the interesting part. A spread from 30 to 840 hours is not a performance distribution; it is two different industries sharing a sample. And a 90-hour mean quietly excludes the forwarders who never replied — arguably the worst outcome in the dataset.
Three constraints travel with that figure and are almost always stripped off when it is quoted:
Nothing did.
More recent figures do circulate. A 2025 post from Rippey AI, a freight automation vendor, reports average website response times of 4–6 hours and email responses ranging from 2 to 46 hours. I was unable to verify its methodology — the page returns no readable content to independent checking tools, and the figures appear without a stated sample size, mode, geography, or definition of what counts as a "response."
That last omission is the fatal one. An automated acknowledgement in four hours and a complete, margin-approved, capacity-checked quote in four hours are wildly different operational achievements. Without knowing which was measured, a 4–6 hour figure cannot be compared to the 2015 result, or to anything else.
The reason no credible benchmark exists is not that nobody has tried. It is that "response time" refers to at least four different events:
A forwarder can acknowledge in seconds and still take three days on step four, because step four may depend on a carrier or co-loader who has not replied yet. Any study that does not say which clock it started and stopped is not measuring something another study can be compared against.
The other statistics that surface in this search are the familiar B2B sales ones — that responding within five minutes makes a lead dramatically more likely to convert.
The underlying research is real. The Short Life of Online Sales Leads, by James Oldroyd and Kristina McElheran, appeared in Harvard Business Review in March 2011, and argued that most companies respond to online inquiries far too slowly.
Two problems with lifting it into freight.
It measures contacting and qualifying inbound sales leads. A freight quote requires lane research, carrier or co-loader pricing, equipment and capacity checks, accessorial validation, and a commercial decision on margin. A sales callback requires a phone.
The specific figures circulating online are frequently not from the HBR article at all. They come from a separate study, and they are typically cited to content-marketing pages rather than to either original. I could not verify the exact numbers against the primary source, which is paywalled. So I am not repeating them here.
That is a small illustration of the larger problem: a statistic gets separated from its methodology, repeated until it sounds like common knowledge, and ends up cited to whoever blogged it most recently.
This is the part I found genuinely surprising. The following questions have no published answer anywhere I could find:
Every one of those is answerable. None has been answered publicly.
It would need timestamped RFQ records pulled directly from forwarders' TMS, CRM, or mail systems — not a survey of what people believe their response times are.
It would define its start and stop events before collecting anything, and publish those definitions. It would include unanswered requests rather than quietly dropping them. It would report medians and percentile bands instead of a single mean. And it would separate inbound website leads, existing-account requests, formal procurement tenders, and direct email, because those carry very different urgency.
The 30-to-840-hour spread in the 2015 data shows the problem plainly. A mean is dragged upward by a handful of very late responses, while dropping non-responders makes the industry look better than it is. Medians with percentile bands, reported alongside the non-response rate, describe what is actually happening.
That is a demanding specification. It is also the only kind of study that would settle the question.
You do not need an industry benchmark to know whether you are slow. Log ten timestamps per RFQ:
Then report the median time to a complete quote, segmented by mode and complexity, alongside your quote-to-booking rate by response-time bucket. Within a quarter you will know more about your own performance than the public literature knows about the entire industry.
Run it for a month before changing anything. Most desks discover the delay is not where they assumed — the wait is frequently on an inbound carrier or agent reply, not on the operator. Automating the wrong step buys nothing.
I am not claiming forwarders are slow. The only real measurement is eleven years old and covers 16 companies on one trade lane.
I am not claiming an hour of delay costs a fixed share of win rate. Nobody has demonstrated that in freight, and anyone quoting a precise figure is extrapolating from sales-lead research conducted in a different industry.
What I am claiming is narrower and, I think, harder to argue with: the metric that every forwarding desk is judged on has not been credibly measured this decade.
We are building this benchmark properly, and we would rather do it with the industry than about it.
Disclosure: I run Zavin, which automates freight quoting — so we have an obvious interest in how this is measured, and we would be a poor choice to mark our own homework. That is exactly why the methodology needs to be public and the data independent.
If you run a forwarding desk and would contribute anonymised, timestamped RFQ data, or if you are a trade publication interested in co-publishing the survey, get in touch. We will publish the methodology in full, report medians rather than flattering averages, include the RFQs that never got quoted, and release the aggregate findings publicly whether or not they suit us.
The 2015 study deserved a successor a decade ago. Our measurement methodology is published at /benchmarks.
There is no credible current benchmark. The most-cited figure — an average of 90 hours — comes from a 2015 study by Freightos that tested 16 forwarders on LCL shipments from China. No comparable study has been published since, and no public source separates response times by mode, forwarder size, or shipment complexity.
From research by Freightos, a freight booking software company, reported by FreightWaves on 11 June 2015. The study submitted online quote requests for less-than-containerload shipments from China to 16 global forwarders. It found an average of 90 hours to receive a quote, with only 45% of forwarders ultimately quoting at all.
It is a reasonable hypothesis, but no public freight study has measured it. Research on general B2B sales leads has long argued that faster response improves contact and qualification rates, but those studies measure sales leads rather than freight bookings, and a freight quote requires carrier pricing, capacity checks, and margin decisions that a sales callback does not.
Mainly because 'response time' has no agreed definition. It can mean an automated acknowledgement, a first human reply, a preliminary indication, or a complete bookable quote — each with a different clock. Without a shared definition and access to timestamped RFQ data, published figures are not comparable.
Median time from RFQ receipt to a complete, commercially valid quote, segmented by mode and shipment complexity, alongside the percentage of RFQs that never receive a quote. Medians and percentile bands are more useful than averages, because a small number of very late responses distorts the mean.
Last updated: August 2026 | v1.0
