Metrics
Dialer report: answer rate, contact rate and cost per answered call
11 November 2026 · 9 min read
The short answer
A dialer report shows how a calling campaign went: how many calls were answered, by whom and at what cost. SautiDesk has four, in the Campaigns group of Reports: Campaign summary, Latest tries, Answer rate by number and Agents on campaigns. Answer rate is answered calls divided by all tries. They are on SautiDesk Pro at KES 3,000 per person a month.
A dialer report answers four questions about an outbound campaign: how many calls were answered, who answered, what it cost and which caller ID number is working. SautiDesk has four campaign reports, one for each question, in the Campaigns group of Reports. They are on SautiDesk Pro, and they cover all three campaign types: pre-recorded, progressive and predictive. This guide gives every figure as the report calculates it, so a supervisor can read a campaign in a few minutes.
Which dialer reports does SautiDesk have?
| Report | The question it answers | Compare to an earlier period |
|---|---|---|
| Campaign summary | How did the calls end, who answered, what did it cost and where do the contacts stand? | Yes |
| Latest tries | What happened on each single call attempt? | No |
| Answer rate by number | Which caller ID number is getting answered and which is failing? | No |
| Agents on campaigns | Which agents took campaign calls and how did they wrap them up? | No |
The four reports appear once the workspace has a campaign of any status. Reports are on SautiDesk Pro and open on a computer. Anyone with the Reports access can read them, and so can anyone with the Campaigns access. For the overall routine across all 36 reports, see the call centre reports guide.
How do the campaign switcher and the filters work?
All four reports share three filters in the header. The campaign switcher lists every campaign in the workspace, finished ones too, newest first, and leaving it empty shows all campaigns together. The Status filter offers All, Running, Paused, Budget reached, Finished and Not started. The Mode filter offers All, Pre-recorded, Progressive and Predictive. Use Mode to compare like with like, because a pre-recorded reminder and a predictive sales list answer very differently.
Dates work as in every report: Today, Yesterday, Last 7 days, This month, Last 30 days or a custom range of up to 366 days. Each report has Export CSV, and Subscribe sends it by email daily, weekly or monthly. The email covers every campaign and does not carry the campaign, status or mode you had selected.
Campaign summary: what each figure means
Campaign summary is the one to open first. Its four headline figures are below, and a table called “How the calls ended” breaks the total down line by line.
| Figure | How it is worked out |
|---|---|
| Calls | Tries made in the range, with the number answered shown under it |
| Answer rate | Answered ÷ all tries. Shown with “N of N calls” |
| Average wait | How long callers waited for an agent, averaged over calls a person answered |
| Cost per answered call | Phone calls + answering-machine checks + AI agent spend, divided by answered calls |
“How the calls ended” lists Calls made, Answered by a person, Answered by an answering machine, Handled by agents, Handed to the AI agent, Hung up while waiting, No answer and Retries (a second try or later), each with its share of calls. Three more tables follow: Contacts, which shows where each called contact stands now, not on the day; Spend, split into Phone calls, Answering-machine checks and AI agent; and By day, with calls, answered, answer rate and spend for each day.
| Line | Figure | How it is worked out |
|---|---|---|
| Calls made | 500 | Tries in the range |
| Answered | 200 | Answer rate: 200 ÷ 500 = 40% |
| Answered by an answering machine | 40 | The other 160 were people |
| Contact rate | 32% | 160 people ÷ 500 calls |
| Spend | KES 900 | Phone KES 800 + machine checks KES 100 |
| Cost per answered call | KES 4.50 | KES 900 ÷ 200 answered |
Read the two percentages together. A 40% answer rate with a 32% contact rate means one answered call in five went to a machine. A campaign can show a good answer rate and still reach few people if the list is full of voicemail, so contact rate is the figure that tells you how many conversations you had.
How do the three campaign types change what you read?
- Pre-recorded. No agent is needed, so read Answer rate, Answered by a person and Cost per answered call. The agent lines matter less here.
- Progressive. One call is placed for each free agent, so waiting should be rare. Check Average wait and Hung up while waiting to see that it held.
- Predictive. More numbers are dialled than there are free agents, and the dialler slows itself down if answered callers start to wait. Average wait and Hung up while waiting are how you see whether it kept that promise. If they rise, the campaign setup is where to ease the pace.
Latest tries: check one contact or one hour
Latest tries lists each call attempt, newest first. Its columns are When, Contact, Number, Campaign (shown only when you are looking at all campaigns), Result, Try, Handled by, Caller ID and Cost. Result reads Answered, Answering machine, Busy, No answer, Failed or In progress. Try is the attempt number for that contact, so a 3 means the third call. Handled by names the first agent on the call, or “AI agent” when AI time was billed.
Use it to settle a question the totals cannot: a customer says nobody called, or the summary dipped at one hour. The screen shows the latest 5,000 tries and the CSV holds up to 50,000, so for a busy campaign narrow the dates first. This is the campaign report to export when someone wants the full list.
Answer rate by number: which caller ID is failing?
A campaign can dial from several caller ID numbers, and customers answer some numbers more than others. Answer rate by number shows three headline figures: Numbers used, Answer rate for the finished calls, and Failing numbers, which counts numbers failing more than 30% of calls. The table lists each number with its label, Status, Tries, Answered, Answer rate and Failed, and has a sort control.
| Status | When a number gets it |
|---|---|
| No status | Fewer than 10 finished tries from that number |
| Good | 10 or more finished tries and a failed share under 20% |
| Watch | A failed share of 20% or more |
| Failing | More than 30% failed, with at least 20 finished tries |
Failed here is calls that failed plus calls that came back busy, divided by finished calls. A call nobody picked up lowers the answer rate but does not count as failed. That difference tells you what to do:
- Low answer rate, low Failed. The line connects and people are not picking up. Look at the hours you dial and the quality of the list, not the number.
- High Failed. The line is not connecting. Test it with a call of your own, and take it off the campaign if the failures continue.
- One number far below the others on the same list. Something is wrong with that number. Keep the rest and drop that one.
When any number is failing, the report says so in plain words and adds that the campaign sends fewer calls through a failing number. Numbers you have taken off a single campaign’s caller ID list still appear with a Removed tag, and tries made before the campaign recorded its caller ID show “Not recorded”. Here is how it reads, using assumed numbers:
| Number | Finished tries | Answered | Answer rate | Failed + busy | Failed | Status |
|---|---|---|---|---|---|---|
| Number A | 100 | 60 | 60% | 10 | 10% | Good |
| Number B | 40 | 8 | 20% | 16 | 40% | Failing |
Agents on campaigns: compare agents fairly
Agents on campaigns covers progressive and predictive calls that an agent took. The headline figures are Agents (who took a campaign call), Calls handled, Talk time and Average call. The By agent table shows Calls handled, Talk time, Average call, Wrapped up and Most common outcome. Two more tables list the Wrap-up outcomes with their share, and the Outcomes by agent.
Compare agents on Calls handled and Average call together, and then on the outcomes they picked. An agent with fewer, longer calls is not behind an agent with many short ones if the longer calls end in the outcomes you want. Calls an agent has not yet wrapped up have no outcome and are left out of the outcome tables, so a low Wrapped up count is a wrap-up habit to coach, not a result to read. Calls the AI agent took are not here and are counted in Campaign summary. For the same fair reading across all calls, see agent scorecard, occupancy and floor standings.
A weekly routine for a campaign supervisor
- Open Campaign summary for Last 7 days with compare on. Note answer rate, contact rate and cost per answered call against the week before.
- Open Answer rate by number. Take any Failing number off the campaign.
- Check Hung up while waiting on progressive and predictive campaigns. If it rises, slow the pace.
- Open Agents on campaigns. Coach the agents with the most unwrapped calls.
- Subscribe to Campaign summary weekly, so the numbers reach your manager without you.
Check the next morning’s missed calls too, because a person you called may ring the number back after the campaign has finished. What the campaign costs is set before you start: outgoing calls are KES 0.05 a second and answering-machine detection is KES 0.50 for each detection, from your prepaid balance.
Start on SautiDesk
- Choose SautiDesk Pro at KES 3,000 per person a month. Campaigns and Reports are both on Pro.
- Add your people and a Kenyan number at KES 500 a month, and top up your balance by M-Pesa or card.
- Follow the nine setup steps and start a small campaign first.
- Open Reports, then Campaigns, after the first hour and again the next morning.
- Subscribe to Campaign summary, daily or weekly.
See the pricing page for every rate, the calling campaigns page for the three types, and all the reports a supervisor gets on Pro.