Key takeaways
- Track three rates, not one: inquiry to quote, quote to order and inquiry to order. Each points to a different fix.
- The denominator is the hard part. Log every inquiry, including phone calls nobody wrote down; Google no longer keeps call history in your Business Profile[2].
- Record channel, source and minutes to first reply on every row, and tag the links you control with campaign parameters[3].
- Use a fixed list of lost-reason codes and judge rates over eight weeks, not one.
- Every Monday: inquiry to order for the trailing eight weeks, quotes gone quiet, and catering calls with no row in the log.
Three rates, three different problems
Ask an owner, a catering manager and a bookkeeper for the catering close rate and you can get three different numbers, all of them defensible. One counts booked orders against quotes sent. One counts them against every inquiry. One counts dollars instead of orders. The disagreement hides the useful part: each version points to a different problem.
| Rate | Formula | A low number usually means |
|---|---|---|
| Inquiry to quote | Quotes sent ÷ inquiries received | Slow replies, missing details, or inquiries that were never a fit |
| Quote to order | Orders booked ÷ quotes sent | Price, menu, follow-up, or a competitor who answered first |
| Inquiry to order | Orders booked ÷ inquiries received | The overall health of catering sales |
| Booked value rate | Booked order value ÷ quoted value | Large jobs slipping away while small ones close |
Example: in a month, a restaurant logs 60 catering inquiries, sends 42 quotes and books 15 orders. Inquiry to quote is 70%, quote to order is about 36%, and inquiry to order is 25%. If the owner only watches quote to order, the 18 inquiries that never got a quote stay invisible, and those are often the cheapest ones to fix.
Keep catering separate from private events in your own dining room. The buyers, lead times and lost reasons differ, and mixing them blurs both. Tracking your event sales pipeline covers the in-house side with its own stages.
Count the inquiries nobody wrote down
Every rate depends on the denominator, and the denominator leaks before anyone sees it. Forms and emails leave a record. Phone calls do not, unless someone writes them down. The lunch-rush call that ended with “someone will call you back” never becomes a row, so it never counts as lost.
Do not expect your Google listing to fill the gap. Google switched off call history in Business Profiles on July 31, 2024, so the profile no longer shows a log of calls placed from your listing[2]. Your phone system's call log and your own inquiry sheet are the record now.
Expect duplicates as well. Buyers who fail on one channel try another: in a Cornell study, 58% of diners said they would still try to book through some other channel if a restaurant was missing from their preferred booking site, and only 18% said they would not book[1]. The office manager who could not get through at noon fills in the web form at 12:20. That is one inquiry, not two. Merge the rows and note both channels, because the first channel tried is the one that failed.
Source: Cornell University Center for Hospitality Research (Kimes and Kies), 2012[1]
Show the data
| Item | Value |
|---|---|
| Still try another channel | 58% |
| Not book at all | 18% |
One row per inquiry
A shared spreadsheet is enough to start. What matters is that every inquiry gets exactly one row, from every channel, with the same columns, filled in by whoever took it.
| Column | Example entry | Why it matters |
|---|---|---|
| Received | Tue 12:14 | Response time starts here, not when someone noticed it |
| Channel | Phone, form, email, marketplace, walk-in | Shows where inquiries arrive and where they stall |
| Source | Google listing, repeat client, referral, flyer | Shows which marketing produces orders, not just inquiries |
| Event date and lead time | Fri 21st, 10 days out | Short-notice requests close differently |
| Headcount and style | 40, drop-off buffet | Lets you compare like with like |
| First personal reply | Tue 14:40 | Auto-replies do not count |
| Quote sent and value | Wed, $880 | Feeds quote to order and booked value |
| Outcome and lost reason | Lost: inside lead time | Turns losses into a list of fixes |
| Owner | Dana | Every open row has a name on it |
Make channel, source and lost reason drop-down lists, not free text. “Google”, “google maps” and “GMB” in the same column turn a five-minute count into an afternoon of cleanup. If you already use catering software, check that it records inquiries that never became quotes; plenty of tools only start counting at the quote, which quietly drops the first rate in the table above.
Get the source right
Channel is how the inquiry arrived. Source is why the buyer found you. A phone call can come from your Google listing, a flyer at an office park or a client's recommendation, and only the source tells you where next quarter's marketing money should go.
- Ask every caller, and put the same question on the form: “How did you hear about us?” with a short fixed list and an “other” box.
- Tag the links you control. Google Analytics reads campaign parameters added to a link, such as utm_source for the referrer, utm_medium for the channel and utm_campaign for the promotion[3]. Tag the catering link in your email signature, on your Google listing and in any marketplace profile, so web inquiries arrive with their source attached.
- Give printed pieces their own route back: a dedicated number or a short web address on the office-park flyer. Call tracking numbers covers when a separate number is worth it.
- Record repeat clients as a source of their own. They close at a different rate, and mixing them in flatters every other channel.
Response time belongs in the log
Record the minutes from inquiry to first personal reply on every row, then group the rows into bands: under an hour, same day, next day, later. Once each band holds 20 or more inquiries, compare inquiry-to-order rates across them. Most owners already suspect speed matters. The bands show how much it matters for your buyers and your menu.
The reason it shows up is plain. Planners are paid to solicit bids from several venues and service providers[4], and office managers ordering lunch behave much the same way. A quote that arrives second competes with one the buyer has already half accepted. Response targets by channel and first-reply email templates help move rows into the faster bands.
Lost-reason codes that lead to a fix
“Lost” on its own teaches nothing. Pick eight or so codes, attach a likely fix to each, and make choosing one a condition of closing the row.
| Lost reason | What it usually means | Where to look |
|---|---|---|
| Never reached | Missed call or unanswered form; no callback connected | Phone coverage and reply speed |
| Inside lead time | Asked for a date the kitchen could not make | A short-notice menu, or published cutoffs |
| Below minimum | Group too small for your order minimum | Minimum stated on the catering page |
| Outside delivery area | Address beyond your radius | A pickup option, or a published radius |
| Price | Buyer said the per-person cost was too high | Menu tiers and how fees are presented |
| Went with another caterer | Booked elsewhere before or after your quote | Ask who and why; compare reply speed |
| No reply after quote | Quote sent, follow-ups unanswered | Follow-up timing and channel |
| Event cancelled | The event itself did not happen | Nothing to fix; leave it out of quote to order |
“No reply after quote” is usually the biggest code and the least understood. Phone follow-ups struggle because eight in ten adults say they generally do not answer calls from unknown numbers[5], so follow up by email first and name the time you will call. Keep those follow-ups about the quote itself: under CAN-SPAM, a message whose content is only commercial has a commercial primary purpose, and opt-out rules come with it[6]. The cadence is in how to follow up on catering quotes.
Small numbers swing hard
Catering volumes are small, so the rates jump around. Example: with 14 quotes in a month, one extra booking moves quote to order by about seven points. A great month followed by a poor one is usually noise, not a trend, and reacting to it with a price change does more harm than the noise did.
- Judge rates over a trailing eight weeks, and compare the same weeks of last year once you have them.
- Do not split by source and by response band at the same time until you have a few hundred rows. The cells get too small to mean anything.
- Watch counts beside rates. A quote-to-order rate that rises because you quoted fewer, safer jobs is not progress.
- Note one-offs in the log, such as a holiday rush or a menu price change, so a jump can be explained later.
Example: what eight weeks of rows can show
Example: a restaurant with a drop-off lunch business logs 96 catering inquiries over eight weeks. Repeat clients send 30 and book 24. New buyers send 66 and book 14. Overall inquiry to order looks healthy at about 40%, but for new buyers it is about 21%.
The 52 new-buyer losses break down as 19 never reached, 11 no reply after quote, 9 price, 7 inside lead time and 6 other. Read that way, the first fix is obvious, and it is not the price. Never reached is the largest code, and every one of those buyers wanted something from the restaurant. The owner adds overflow coverage at lunch and a host catering card, then reads the same numbers eight weeks later. Price, the code everyone argues about, comes fourth.
The repeat-client split matters just as much. In this example, without it, the 40% overall rate would hide that new buyers convert at about half that, and the restaurant would keep paying for marketing that brings in inquiries it then fails to answer.
Who keeps the log
A log that three people half-maintain is worse than none, because it looks complete when it is not. Give it one owner, usually the catering manager, and a few rules for everyone else.
- Whoever takes an inquiry creates the row before the end of their shift, even if it only holds a name, a number and a date.
- Only the owner closes rows and picks lost-reason codes, so a code means the same thing every week.
- Managers covering the catering phone on the owner's day off use the same drop-down lists, not notes in the margin.
- Rows are never deleted. A duplicate is merged into the first row, with both channels noted.
How an inquiry moves through the log
- Caller
- Your team
01Caller
Inquiry arrives
Call, form or email; the row starts with the time received.
02Your team
Row created
Channel, source, date, headcount and style entered; duplicates merged.
03Your team
First reply logged
Minutes to a personal reply recorded. Auto-replies do not count.
04Your team
Quote sent
Date and value entered, follow-up dates set on the row.
05Your team
Closed won or lost
Booked value, or one lost-reason code from the fixed list.
Example flow. A row cannot close without an outcome and, if lost, a reason.
Phone inquiries are where rows go missing, so automate that step where you can. An SI Employee answering your forwarded catering calls on overflow or after hours puts each call's summary, recording and transcript in your AssistCall inbox and emails the person you choose. Signed webhooks can also post each summary to a spreadsheet or CRM through tools like Zapier, so a phone lead becomes a row without anyone retyping it; integrations lists the routes. For the first few weeks, have whoever owns the log check each new row against the call summary.
Three numbers for Monday
A weekly review keeps the log honest. Fifteen minutes, the same three numbers, in the same order, before anyone talks about menus or prices.
Inquiry to order
Trailing eight weeks, overall and by source, with the counts beside each rate
Quotes gone quiet
Quotes past their follow-up date with no contact logged
Unlogged calls
Missed calls in catering hours with no matching row in the log
Review these first. The third number tells you whether the first two can be trusted.
Close last week's rows
Mark anything won or lost with an outcome and, for losses, one reason code.
Read the three numbers
Inquiry to order for the trailing eight weeks, quotes gone quiet, unlogged calls.
Assign the quiet quotes
Each one gets an owner and a follow-up today, by email first.
Match the call log
Compare missed calls in catering hours with new rows. Add what you can identify and call those numbers back from the restaurant line.
Pick one fix
Take last month's most common lost reason and choose a single change to test for four weeks.
Put inquiry to order on your weekly restaurant dashboard next to sales and labor, so it gets the same attention. To put a dollar value on the unlogged calls, use the calculator in missed catering calls: the revenue at risk. The restaurant group catering example shows one way phone leads could reach the tracker across several locations.
Frequently asked questions
What is a good catering close rate?
Published benchmarks for restaurant catering are scarce, and rates swing with menu, price point, lead source and how many inquiries were never a fit. Track your own inquiry-to-order and quote-to-order rates over a trailing eight weeks and work on the trend.
Should cancelled events count as lost?
No. Give them their own code and leave them out of quote to order. Otherwise a cancelled event looks like a pricing or follow-up problem.
How do I track phone inquiries if nobody writes them down?
Give hosts a short catering card that becomes a log row, compare the phone system's call log with the inquiry log every Monday, and consider answering overflow calls with a service that sends a summary of every call.
Do repeat clients belong in the close rate?
Track them, but as their own source. They close far more often than new buyers, and mixing them in hides how well new inquiries convert.
Can Google tell me how many people called from my listing?
Business Profile no longer keeps call history; Google switched it off on July 31, 2024[2]. Use your phone system's log and your own inquiry sheet instead.
References
Cornell University Center for Hospitality Research (Kimes and Kies) (2012). The Role of Multi-Restaurant Reservation Sites in Restaurant Distribution Management (Cornell Hospitality Report, 2012) (opens in a new tab) ↑
Google Business Profile Help (2024). Changes to Google Business Profile chat and call history (opens in a new tab) ↑
Google (Analytics Help) (2026). URL builders: Collect campaign data with custom URLs (opens in a new tab) ↑
US Bureau of Labor Statistics (2026). Occupational Outlook Handbook: Meeting, Convention, and Event Planners (opens in a new tab) ↑
Pew Research Center (2020). Most Americans don't answer cellphone calls from unknown numbers (opens in a new tab) ↑
Federal Trade Commission (2026). CAN-SPAM Act: A Compliance Guide for Business (opens in a new tab) ↑