---
title: "962 Calls in Two Months: What Real Trade Call Data Shows (2026)"
description: "We analysed 962 calls handled for three home service businesses across May and June 2026. What the data shows about urgency, call length and what actually gets missed."
canonical: https://dolfyn.ai/call-data
source: dolfyn.ai
---

# 962 Calls in Two Months: What Real Trade Call Data Shows (2026)

> We analysed 962 calls handled for three home service businesses across May and June 2026. What the data shows about urgency, call length and what actually gets missed.

# 962 calls in two months, and what they show about what actually gets missed

 
Most statistics about missed calls in the trades come from vendor surveys. These come from call logs. We looked at every call three home service businesses handled through dolfyn across May and June 2026, two complete months.

 
962 calls · 3 businesses · May and June 2026 · all customer details removed


 

962

calls handled in two months across three businesses


 

38%

of triaged calls were urgent, not routine


 

58s

median call length


 

93%

of calls resolved in under three minutes


## Why this is different from the usual numbers


The figures you normally see quoted about missed calls come from surveys, vendor blogs, or panels that are rarely published in full. We have repeatedly found the same statistic attributed to four different sources with four different values.


This is not that. This is what three real businesses actually received: an HVAC and plumbing company in British Columbia that uses the agent for daytime overflow and after-hours cover, a heating and plumbing company in Saskatchewan that routes every call through it, and a multi-trade contractor in Ontario that uses it for daytime overflow only. Every caller name, phone number, address and email has been removed. What is left is the shape of the demand.


## Nearly half of triaged calls were genuine emergencies


Across the calls where the agent classified urgency, **38% were urgent rather than routine**. That figure is not evenly distributed, and the way it splits is the most useful thing in this data. The British Columbia company uses the agent for daytime overflow and after-hours cover: **42% of its triaged calls were urgent**. The Ontario contractor uses it for daytime overflow only: **32% were urgent**.


Both businesses are forwarding the same kind of call: the one nobody could get to. The only real difference between them is that one of them also covers evenings and weekends, and that single difference lifts the share of emergencies by roughly a third. Out-of-hours calls are not simply more calls. They are a different mix, weighted toward the work that cannot wait until Monday.


## What people were actually calling about


 CategoryShare of categorised calls
 Plumbing26%
 Air conditioning18%
 Heating14%
 HVAC, general6%
 Electrical1%
 Other, including billing, scheduling and follow-ups35%


That last row matters more than it looks. Around a third of calls were not new work at all. They were existing customers rescheduling, asking about an invoice, chasing a technician, or following up on a quote. Those calls still have to be answered, and they are the ones a busy owner is most likely to let ring.


## Most calls are short


The median call ran **fifty-eight seconds**, and **93% finished inside three minutes**. An emergency intake takes longer, but the bulk of the volume is quick: confirm who is calling, what they need, and when someone can be there.


This is the part that makes missed calls feel deceptively cheap. Each individual call looks like a minor interruption. In aggregate, across two months, it was 962 of them.


## How callers sounded


Sentiment was scored on every call: **87% neutral, 8% positive, 4% negative**. The negative share is worth sitting with rather than dismissing. Roughly one call in twenty involved a frustrated caller, which is what you would expect from a stream weighted toward emergencies and follow-ups.


## What we are deliberately not claiming


 
Three things we could have published and chose not to.

 
**A single after-hours percentage.** The three businesses deploy the agent differently: one covers daytime overflow plus evenings and weekends, one takes every call, one handles daytime overflow only. Blending them into a single number would produce a statistic that describes nothing real.

 
**Precise hour-by-hour timing.** The exports do not carry consistent timezone information across accounts, so we are not publishing a peak-hour chart we cannot stand behind.

 
**Revenue recovered.** We know what the calls were. We do not have each business's close rate or job values, so any dollar figure would be a guess wearing a decimal point.


## What the data shows


Three businesses, two months, 962 calls that would otherwise have rung out or gone to voicemail. Nearly two in five of the triaged ones could not wait, and that share climbs once evenings and weekends are included. A third were not new leads at all but existing customers who still needed an answer. Almost all of it was handled in under three minutes.


The argument for answering every call has never really been about technology. It is that the calls you miss are not a random sample of the calls you get.


 
## See what your own call data looks like

 
Two weeks, no credit card. At the end of it you will have your own version of this page, built from your actual calls.

 [Book a demo](https://dolfyn.ai/#demo)

---
*Source: [https://dolfyn.ai/call-data](https://dolfyn.ai/call-data)*
*dolfyn — AI voice receptionist for contractors and service businesses*
*Starts at $179/month. 2-week free trial. No contracts. [dolfyn.ai](https://dolfyn.ai)*