Phones and call handling

Measuring phone performance across a restaurant group

A monthly phone report for multi-site restaurant and pub groups: call volume, answer rate, booking conversion and peak hours, compared fairly by site.

Ava research teamPublished 7 min read

On this page
  1. Why a group needs one phone report
  2. The seven numbers to track
  3. Getting the data out
  4. Check the data before you trust it
  5. Compare sites fairly
  6. What the patterns usually mean
  7. Put a value on time spent on the phone
  8. A one-page monthly report
  9. Example: three sites, one month
  10. Set targets you can keep
  11. Frequently asked questions
  12. References

Key takeaways

  • Track seven numbers per site each month: volume, answer rate, missed calls at peak, booking conversion, after-hours share, handovers and callback time.
  • Compare sites per 100 covers and by opening hours, or the busiest site will always look worst.
  • Join call records to booking data by source, so you can see which missed calls cost covers.
  • Use the report to change rotas and websites, not to blame individual sites.

Why a group needs one phone report

Multi-site operators are a big part of UK hospitality. Restaurants and mobile food service businesses ran 125,660 local units in 2026, against 108,500 enterprises[1], so a large number of sites sit inside businesses with more than one. Yet phone performance is usually managed site by site, if at all. Each general manager has a sense of whether the phone is a problem; nobody can see the whole picture.

The managed groups that report trading figures show why small gains matter. Like-for-like sales at Britain's leading managed hospitality groups were 2.7% higher in September 2026 than a year earlier, their best growth of the year[2]. Managed restaurant groups grew by 2.6% and pub groups by 3.8%[3][4], while managed bar groups fell 1.7%[5]. When annual growth is measured in low single figures, the covers lost to unanswered phones at one or two weak sites can be the difference between a good year and a flat one.

Like-for-like sales growth at managed hospitality groups, September 2026Year on year; managed bar groups fell 1.7% and are not shown
Like-for-like sales growth at managed hospitality groups, September 2026Bar chart. Managed pub groups: 3.8%; All managed groups: 2.7%; Managed restaurant groups: 2.6%.Managed pub groupsManaged pub groups: 3.8%3.8%All managed groupsAll managed groups: 2.7%2.7%Managed restaurant groupsManaged restaurant groups: 2.6%2.6%Like-for-like sales growth at managed hospitality groups, September 2026Bar chart. Managed pub groups: 3.8%; All managed groups: 2.7%; Managed restaurant groups: 2.6%.Managed pub groupsManaged pub groups: 3.8%3.8%All managed groupsAll managed groups: 2.7%2.7%Managed restaurant groupsManaged restaurant groups: 2.6%2.6%

Source: NIQ (CGA by NIQ) with RSM, 2026[2]

Show the data
ItemValue
Managed pub groups3.8%
All managed groups2.7%
Managed restaurant groups2.6%

The seven numbers to track

Seven phone metrics for a monthly group report
MetricDefinitionWhy it matters
Call volumeInbound calls per site per month, excluding internal and supplier linesShows demand and the load on front of house
Answer rateCalls answered by a person or assistant, divided by all callsThe headline measure of whether guests get through
Peak missed callsUnanswered calls in the two busiest hours of each dayWhere lost bookings concentrate
Booking conversionPhone bookings divided by calls about bookingsShows whether answered calls turn into covers
After-hours shareCalls outside opening or staffed hours, as a share of all callsPoints to website and greeting gaps
HandoversCalls passed to a manager, by reasonShows load on managers and recurring problems
Callback timeAverage time from a missed call or handover to the return callGuests remember slow callbacks

Resist adding more. A report with twenty measures gets skimmed; one with seven gets discussed. If a site's numbers raise a question, dig into the detail for that site rather than adding columns for everyone.

Getting the data out

  1. Phone system call records

    Most business phone systems and VoIP providers can export a call log with date, time, duration, the number dialled and whether the call was answered. Ask for a monthly export per site in the same format.

  2. Booking system by source

    Export bookings and covers by channel: phone, website, Google, third-party apps and walk-in. Make sure staff select the right source when they enter a phone booking.

  3. EPOS covers

    Take total covers per site from the EPOS so you can express phone figures per 100 covers.

  4. Assistant logs

    If an assistant such as Ava answers overflow or after-hours calls, include its call log and handovers, so the report counts every call once.

  5. Opening hours

    Keep a simple table of each site's opening and phone-staffed hours for the month, including closures.

The first month is the hardest. Once the exports are set up, a spreadsheet with one tab per source and a summary tab can be refreshed in under an hour.

Check the data before you trust it

Phone reports go wrong in predictable ways. A few checks in the first two months save arguments later, and they make general managers far more willing to act on what the report shows.

  • Count the lines. Make sure every published number for each site, including old numbers still listed on directories, is in the export.
  • Strip out repeat calls. One guest ringing four times in five minutes is one missed guest, not four, so count unique callers per hour as well as calls.
  • Check booking source discipline. Watch a site enter a week of bookings; phone bookings mislabelled as walk-ins make conversion look worse than it is.
  • Match the clocks. The phone system, booking system and EPOS should all use UK time, including the switch to and from British Summer Time.
  • Spot-check five calls per site. Compare the log with what staff remember, especially calls marked answered that lasted only a few seconds.

Once the numbers survive these checks, managers start trusting the report, and that is the point at which it starts changing rotas and habits rather than sitting in an inbox.

Compare sites fairly

Raw numbers punish success. The busiest site will always have the most missed calls, and a site that closes on Mondays will have fewer calls than one that opens seven days. Normalise before comparing.

  • Express volume and missed calls per 100 covers, so a 60-cover pub and a 200-cover restaurant sit on the same scale.
  • Split calls into staffed hours and after hours; judge answer rate only on staffed hours, and track after-hours calls separately.
  • Compare each site with its own previous three months as well as with the group, because trend matters more than rank.
  • Note one-off events, such as a refit, a local festival or a big match, that will distort a month.
  • Group sites by format, for example city restaurants, suburban pubs and travel locations, before ranking them.

Our separate guide on comparing sites fairly goes further on adjusting for size, location and format across all metrics, not only phones.

What the patterns usually mean

  • Low answer rate concentrated between 6pm and 8pm: nobody owns the phone during the first sitting. Fix the rota or add overflow cover.
  • Good answer rate but low booking conversion: callers are being told no. Check availability rules, party limits and whether staff offer alternatives.
  • High after-hours share: the website or Google listing is not answering basic questions. Check hours, menus and the booking link.
  • Rising handovers for one reason, such as allergens: the menu information or staff training needs attention.
  • Slow callbacks: missed calls are not being checked. Set a fixed time each shift to return them.
  • One site with far more calls per 100 covers: its listing may show the wrong hours or the wrong number.

Put a value on time spent on the phone

Phone time is staff time. Median hourly pay for waiters and waitresses in the UK was £12.81 in April 2026, excluding overtime[6], and that is before employer costs. Example: a site taking 900 calls a month at an average of three minutes each spends 45 hours a month on the phone, more than a full week of one person's shifts. Cutting a third of those calls through better online information frees around 15 hours for the floor.

Show this figure in the report alongside the call volume. It turns an abstract measure into something a general manager can weigh against other uses of the team's time.

A one-page monthly report

  1. Headline: group answer rate in staffed hours, and the change from last month.
  2. Site table: the seven metrics per site, normalised, with the best and weakest shown against each site's own trend.
  3. Peak heatmap: missed calls by hour and day for the group, to show where the gaps are.
  4. Top three reasons for handovers, with the count by site.
  5. Actions agreed last month, and whether they were done.
  6. Two or three actions for next month, each with an owner.

Keep the tone practical. The report exists to fix rotas, websites and processes, not to rank general managers. Sites that feel judged stop recording phone bookings accurately, and the data gets worse.

Example: three sites, one month

Example: a fictional group runs three sites. The city restaurant takes 1,400 calls, answers 78% in staffed hours and has 160 missed calls at peak. The suburban pub takes 700 calls, answers 91% and misses 25 at peak. The station café takes 300 calls, answers 85%, but 40% of its calls arrive after hours.

Example: per 100 covers, the city site's missed calls are three times the pub's, all between 6pm and 8pm on Thursday to Saturday. The group adds a named phone owner for those hours and an overflow option. The café's after-hours calls are mostly about opening times on Sundays, which its Google listing showed wrongly. One correction and a rota change later, the next month's report shows the city answer rate at 88% and the café's after-hours share halved.

Set targets you can keep

Set targets per format rather than one number for everyone, and move them gradually. A busy city site might aim to answer nine in ten calls in staffed hours and return every missed call within two hours; a quieter pub may already be there. Review targets every quarter alongside the handover rules, because both depend on who is on shift. And when visitors make up a big share of calls at a site, add a check on calls in other languages, which often take longer and are more likely to need a written follow-up.

Frequently asked questions

What is a good answer rate for a restaurant phone?

There is no published UK benchmark. Set your own by format and aim to improve on each site's last three months. Judge it on staffed hours only, and track after-hours calls separately.

How do we know which missed calls cost bookings?

Compare missed calls at peak with phone bookings for the same hours, and check whether missed callers later booked online. Many phone systems can show whether a missed number called back.

Should supplier and internal calls be counted?

No. Exclude known supplier, staff and internal numbers, or the report will overstate guest demand.

How often should we review the phone report?

Monthly is enough for most groups, with a quick weekly check on missed calls at the busiest sites during peak seasons such as December.

References

  1. Office for National Statistics (2026). UK business: activity, size and location, 2026 (workbook) (opens in a new tab) ↑

  2. NIQ (CGA by NIQ) with RSM (2026). Bright September gives hospitality groups their best growth of 2026 (NIQ RSM Hospitality Business Tracker) (opens in a new tab) ↑

  3. NIQ (CGA by NIQ) with RSM (2026). Bright September gives hospitality groups their best growth of 2026 (restaurant groups) (opens in a new tab) ↑

  4. NIQ (CGA by NIQ) with RSM (2026). Bright September gives hospitality groups their best growth of 2026 (pub groups) (opens in a new tab) ↑

  5. NIQ (CGA by NIQ) with RSM (2026). Bright September gives hospitality groups their best growth of 2026 (bar groups) (opens in a new tab) ↑

  6. Office for National Statistics (2026). ASHE 2026 provisional, Table 14.6a: hourly pay excluding overtime by occupation (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo