The Australian Pharmacy Phone Report: What 3,988 Calls Revealed

Most pharmacy owners know they miss calls. Almost none know how many, when, or who was calling.

So we measured it. Over 30 days in 2025, we logged every inbound call to one independent Australian community pharmacy. That is 3,988 calls. The pharmacy answered 1,001 of them.

This report sets out what we found, how we measured it, what it means, and how to run the same audit on your own phone line. The headline result is not that a pharmacy missed calls. It is when it missed them, and how badly the standard industry assumption about after hours turns out to be wrong.

Key findings: where pharmacy missed calls actually happen

• 74.9 percent of inbound calls went unanswered. That is 2,987 missed calls in 30 days, or roughly 99.6 unanswered calls a day.
• 99.1 percent of calls arrived during opening hours. After hours accounted for less than 1 percent of total call volume.
• Answer rate fell as volume rose. Of the weekdays measured, the busiest had the worst answer rate.
• The average abandoned caller waited 19.2 seconds. Most did not hang up instantly. They waited, then gave up.
• 11am and 1pm were the two busiest hours, with 702 and 699 calls respectively across the month.
• 4pm was the best performing hour at a 35.3 percent answer rate, still well below any reasonable target.

Finding 1: three quarters of calls never got through

Across the 30 day window the pharmacy received 3,988 inbound calls and answered 25.1 percent of them.

The remaining 2,987 calls either rang out or were abandoned by the caller. Averaged across the month, that is 99.6 unanswered calls per day. We count calls rather than people here, because a caller who redials appears more than once.

For context on what that volume represents, Australians visit community pharmacies more than 443 million times a year across more than 6,000 locations. The AIHW records 335.1 million prescriptions dispensed under the PBS and RPBS in 2024-25. Each of those scripts can generate a phone call about timing, stock, interactions, or collection.

The phone is not a secondary channel for a pharmacy. On the evidence of this audit, it is a primary channel operating at roughly a quarter of its capacity.

Finding 2: this is not an after hours problem

This is the finding that surprised us, and it runs against how the whole category is usually sold.

99.1 percent of all inbound calls arrived within the pharmacy’s operating hours. Fewer than 1 in 100 calls came in when the doors were shut.

That matters, because “cover your after hours calls” is the standard pitch for phone automation in this industry, including in some of our own earlier writing. On this data, after hours coverage addresses under 1 percent of the missed call volume. The other 99 percent of the problem happens while the pharmacy is open, staffed, and trading.

The mechanism is capacity, not availability. The phone rings while the same people are serving the counter, checking a script, taking a delivery, or running a consultation. There is nobody free to answer, so it rings out.

Weekends behaved the same way. Weekend days averaged 53.9 calls at a 22.7 percent answer rate, so roughly 42 unanswered calls per weekend day. Lower volume, similar failure rate.

None of this makes after-hours coverage worthless, and we cover that slice in our guide to after-hours pharmacy call coverage. It just is not where the volume is.

Finding 3: answer rate collapses exactly when demand peaks

The intuitive assumption is that a pharmacy misses calls when it is short staffed on a quiet day. The data shows the opposite.

Thursday averaged 213.8 calls a day and answered 19.9 percent of them. That is roughly 171 unanswered calls in a single day.

Friday averaged 109.8 calls, close to half Thursday’s volume, and answered 33.0 percent. Tuesday sat in between on both measures, at 201.2 calls and 29.6 percent.

The pattern is consistent. The more the phone rings, the smaller the share of calls that get picked up. Capacity does not stretch to meet demand, it breaks under it. A pharmacy does not gradually fall behind on a busy day, it falls off a cliff.

The busiest single day in the audit was Monday 2 June, with 318 calls.

Finding 4: callers wait, then give up at around 20 seconds

Of the 2,987 unanswered calls, we can see how long the caller held on before the call ended. We treat these as abandoned calls throughout.

The average abandoned caller waited 19.2 seconds.

That distribution matters more than the average. A common contact centre convention is to treat calls abandoned inside 10 seconds as misdials and discard them. Apply that convention here and 83.9 percent of these abandons were real callers who deliberately waited and then gave up.

In other words, this is not noise. Around 2,507 calls over 30 days involved someone waiting on a ringing phone, deciding nobody was coming, and hanging up.

The 11 to 30 second band is the one worth acting on. It holds 1,206 calls, 40.4 percent of all abandons. Those callers are patient enough to wait through the first several rings but not patient enough to wait through a rush. A response inside the first few seconds captures almost all of them.

Finding 5: the pressure concentrates in two hours

Call volume was not evenly spread across the trading day. Two hours dominated.

11am recorded 702 calls across the month. 1pm recorded 699. Together, those two hours accounted for over a third of total call volume.

4pm was the best performing hour at a 35.3 percent answer rate. It was still the best of a poor set, and even that hour missed roughly two calls in three.

If you have ever wondered why the phone feels relentless mid morning and again straight after lunch, this is why. Those are also the hours when the dispensary is deepest into a queue and the counter is busiest, which is exactly when nobody can reach the handset.

Why this happens, and why it is a safety issue too

The phone competes directly with dispensing, and dispensing loses either way.

A 2025 Monash led scoping review of 51 studies found that pharmacy staff are interrupted as often as every two to six minutes during dispensing. Rates ranged from under 5 to more than 20 interruptions per hour. It was published in Research in Social and Administrative Pharmacy. Phone calls are listed first among the most common external interruptions during dispensing.

The same review notes that research summarised in its background attributes approximately 9.4 percent of dispensing errors to interruptions and distractions. We cover the other contributing conditions in how to reduce dispensing errors.

So the pharmacy faces a genuine two sided problem. Answer the phone and you interrupt the bench, which carries a measurable accuracy cost. Do not answer it, and three quarters of your callers do not get through. Neither option is good, and most pharmacies end up doing both badly at once.

Grattan Institute analysis puts the underlying pace in perspective. Australian pharmacies dispense roughly one prescription every 62 seconds on the time and motion evidence it examined. Grattan notes that figure comes from a small sample and may not be nationally representative.

Demand is also shifting in a direction that loads the phone further. The Pharmacy Guild reports that prescription volumes remain flat following the move to 60 day dispensing, while consultation services grow sharply. More than 80,500 UTI consultations were delivered in 2025, and pharmacists administered 29 percent of all flu vaccines. Consultations are appointment shaped. Appointments are booked by phone.

What a missed call is actually worth

We want to be careful here, because this is where most writing on this topic starts inventing numbers.

There is no published Australian figure for the lifetime value of a community pharmacy customer. We looked. Guild, PSA, ABS, Grattan and the academic literature do not publish one. Anyone quoting you a precise dollar value per missed pharmacy call is estimating, not citing.

What can be said with sources is narrower and more useful:

• Grattan puts pharmacy remuneration at about $15.90 per script in 2024-25 in today’s dollars. That is far closer to the value of a script to the pharmacy than the dispensed price, most of which is medicine cost paid by government.
• Access friction is already high across Australian healthcare. The ABS found 26.0 percent of Australians waited longer than they felt acceptable for a GP appointment in 2024-25. A further 26.6 percent delayed or did not see a GP when needed.
• Reaching healthcare by phone is hard well beyond Australia. In the NHS England GP Patient Survey 2025, run by Ipsos with 702,837 respondents, only 52.9 percent said it was easy to contact their practice by phone. That is UK general practice rather than Australian pharmacy, but it is the largest robust dataset that measures this exact experience.

The honest framing is this. We cannot tell you what one missed call costs. We can tell you that in this pharmacy, roughly 2,987 calls a month went unanswered. Not one of those requests appeared in any report the owner was reading.

That is the real problem with missed calls. They are invisible. A dispensed script is recorded. A call that rang out is not.

What changed when an AI voice agent answered

After the audit, the same pharmacy ran a limited beta of Emily, Krepko’s AI voice agent, later in 2025.

Two numbers are worth reporting, and one comparison is worth refusing.

During active operating periods across the beta, every call routed to Emily was answered. She picked up in under 5 seconds, well inside the window in which 83.9 percent of abandoned callers were still holding. Of the calls Emily handled, 57 percent produced an actionable ticket for staff (112 tickets from 195 calls), and staff completed 84.8 percent of those tickets.

The request mix across those 112 tickets was:

Request type	Tickets	Share
Delivery	45	40.2%
Pickup confirmation	31	27.7%
Prescription inquiry	27	24.1%
Other	9	8.0%

Almost none of it is clinical. It is logistics.

The comparison we will not draw is a volume comparison. The audit covers 3,988 calls across 30 full days. The beta covers 195 calls across 46.8 hours of active operation. Those are different windows of very different sizes, and presenting them as a single before and after funnel would be misleading. Rates are comparable between them. Totals are not.

We have written the operational detail up separately in our guide to AI voice agents and missed calls at Australian pharmacies. The largest single category of request was delivery, which we cover in what call data says about pharmacy delivery demand.

Methodology

What we measured. Every inbound call to the pharmacy’s main published phone line, including answered calls, calls that rang out, and calls abandoned by the caller.

Site. One independent Australian community pharmacy. Single site. The pharmacy has been anonymised.

Window. 30 consecutive days in mid 2025, covering the full trading week including weekends.

Sample. n = 3,988 inbound calls. 1,001 answered, 2,987 unanswered.

Definitions. A call is answered when a staff member picks up. A call is abandoned when the caller disconnects before pickup. Wait time is measured from the start of ringing to disconnection. After hours means outside the pharmacy’s published trading hours on that day.

Beta comparison. The AI voice agent figures come from a separate 30 day window later in 2025, covering 195 calls across 46.8 hours of active operation at the same site.

Scope of the beta figures. The agent ran only during limited active periods, not across the full month of call traffic. At 195 calls over 46.8 hours it saw a fraction of the volume in the baseline audit. So “every call was answered” means every call routed to the agent while it was running. It is not a claim about the whole call load.

Limitations

We would rather state these than have you find them.

• We counted calls, not callers. The audit measures inbound calls. At a 25.1 percent answer rate some people will have redialled, so the number of distinct patients affected is lower than the call count. We cannot deduplicate them from this data.
• This is one pharmacy. It is a real, complete, 30 day census of that pharmacy’s phone line rather than a sample, but it is n = 1 site. It is not nationally representative and we do not claim it is.
• Trading pattern affects everything. A pharmacy inside a medical centre, or one with a large delivery run, will have a different call profile.
• The data is from 2025. Call patterns shift, and pharmacist scope of practice has expanded since. We intend to re-run this audit and update this page.
• Day of week figures are partial. The audit reported daily breakdowns for Tuesday, Thursday, Friday and weekends. We have not estimated the missing days and have not included them.
• We build a product in this category. Krepko sells an AI voice agent for pharmacies, so we have an interest in the finding that pharmacies miss calls. That is precisely why the methodology and limitations are published here, and why we have declined to invent a dollar value per missed call.

How to run this audit on your own phone line

You do not need us, or any vendor, to find out your own numbers. Most Australian pharmacy phone systems and VoIP providers already log this.

1. Pull 30 days of call detail records from your phone provider. Ask for inbound calls with timestamp, duration, and disposition (answered, missed, abandoned).
2. Calculate your answer rate. Answered calls divided by total inbound calls. As a rule of thumb rather than a published standard, we would treat anything under 80 percent as worth acting on.
3. Split by in hours and after hours. If your split looks like ours, stop treating this as an after hours problem.
4. Group abandoned calls by wait time into under 10 seconds, 11 to 30 seconds, and over 30 seconds. The middle band is where the recoverable calls sit.
5. Chart answer rate against volume by day and hour. Look for the inversion. If your worst answer rate lands on your busiest day, you have a capacity ceiling rather than a rostering problem.
6. Do it before you buy anything. Any option you consider, from a second handset to an answering service to an AI agent, should be measured against this baseline.

If you run this and your answer rate is above 80 percent, you do not have a phone problem, and you should spend your money elsewhere.

Where Krepko fits

Krepko builds Emily, an AI voice agent for Australian pharmacies. Emily answers the phone, works from the pharmacy’s own live data rather than a generic script, and passes clinical or complex calls to a pharmacist rather than attempting them. She runs on a private, Australia based server.

We published this report because our own starting assumption turned out to be wrong. We thought missed calls were an after hours problem. In the only dataset we could measure properly, they were not. The problem is in hours capacity, at 11am and at 1pm, on the busiest days.

If you want to understand the options before you talk to anyone, start with our comparison of AI voice agents against voicemail and answering services. You can also read how phone interruptions affect dispensing accuracy at the bench. If you want to talk to us, we are at krepko.com.au.

We intend to re-run this audit across more sites. If you are an Australian pharmacy willing to contribute anonymised call records to the next edition, we would like to hear from you.