When to call again, and when to stop

The short answer

When nobody spoke, the attempt ends in a line result. It tells you nothing about interest, only that the moment did not work. So try again at a new time, up to a cap. When someone answered and told you something, the attempt ends in a spoken answer, and there is nothing to retry. You follow the answer: "Call me later" becomes a callback at their time, "Not interested" stops this campaign, and "Do not call" stops every campaign.

The attempt cap, the most attempts per contact in one campaign, comes from the campaign's purpose. Did the person ask for this call, do they need to know something, or is it an offer they did not ask for? Set the cap from that answer, then check it against the attempt yield of your last campaign with the same purpose, instead of copying it from habit. The attempt yield is the people reached at one attempt divided by the calls placed at that attempt.

In one sentence: retry a line result, follow a spoken answer.

Country rules come first. Calling hours, attempt limits, consent and do-not-call registers differ by country and sometimes by sector, and this policy works inside them. This article is not legal advice. Number reputation and spam labels are a separate topic.

Two kinds of result

An attempt is one call the campaign places to one contact. A callback at a time the person asked for is not an attempt. Every attempt ends in one of two kinds of result:

• Line result: nobody spoke. No answer (it rang and nobody picked up), Busy (the line was in use) or Voicemail (a machine answered).
• Spoken answer: someone answered and told you something. Completed (the call did what it was for), Call me later, Not interested, Wrong number or Do not call.

A person who answers is reached, whatever they then say. A machine is not a person, so Voicemail is a line result.

Line results are retried. A retry after No answer goes to a new time: a different time block or another day. A time block is a part of the calling day: morning, midday, afternoon or early evening. Spoken answers are followed, each with its own next step.

Four of these rows need a closer look.

Busy: never call straight back

Busy can look like the person is right there and will be free in a minute. But on some networks, a call the person rejects also comes back as Busy. A call straight back then rings someone who has just declined your call. So retry Busy in a different time block.

Call me later: a callback, not an attempt

"Call me later" is a spoken answer: the person picked up and wants the call at another time. So the contact leaves the retry pool and gets a callback, a call at the time the person named. It does not count toward the attempt cap, because the person asked for it.

If the person does not name a time, the agent offers two and the person picks one:

• Contact: "I'm about to go into a meeting. Can you call me later?"
• Agent: "Of course. I can call you back today at four, or tomorrow morning at ten. Which works better for you?"
• Contact: "Tomorrow at ten is good."
• Agent: "Great. Then I'll call you tomorrow at ten, on this number. Talk to you then."

The agent offers only times inside the calling window, the hours and days a contact may be called. If the person names a time outside it, the agent offers the closest time inside it. If nobody picks up at the agreed time, the No answer rule applies from there: the next call goes to a new time and counts toward the cap.

Not interested and Do not call: two different stops

Not interested is a no to this purpose. Stop this campaign for that contact, and record the answer together with the purpose, so that no campaign with the same purpose calls again. It does not block a call with another purpose. A person who declined a renewal offer still needs to hear that their delivery is late.

Do not call is a no to every call. Stop every campaign, and store the answer where every campaign reads it, not only in the campaign that heard it. Our guide to a CRM write-back data model shows how to store outcomes like these in your CRM.

Set the attempt cap by purpose

The purpose is why the campaign calls. A person who asked for the call is waiting for it. A person who needs to know something should hear it soon. A person who did not ask for the call has given you no reason to keep trying for long. So the cap and the spread follow the purpose.

This guide uses four purposes. Example policy, not a rule, and always inside your country's rules:

Spread the attempts. A cap of 6 does little if all six calls fall in one afternoon. Take a quote request that comes in on a Monday at 10:00. Here is one schedule that keeps every rule: Monday morning, within 15 minutes, and Monday early evening, then Tuesday midday, Wednesday afternoon, Thursday morning and Friday early evening. That is 6 attempts over 5 working days, never more than 2 a day, and each retry in a different time block from the one before.

Voicemail. Leave at most one message per campaign, and decide per purpose whether a message is allowed at all. A short message that says who called and why suits a Requested contact or a Service notice better than an Unrequested offer. Either way, the contact is then retried as No answer.

Contact ceiling. A cap per campaign is not enough when one person sits in several campaigns. So set a contact ceiling: a limit per contact across all campaigns, per day and per week. In this example policy, it is at most 2 attempts a day and 6 a week. The Requested contact cap of 6 over 5 working days, at most 2 a day, fits inside it. Callbacks are not attempts, so they do not count toward the ceiling either.

Which purpose goes first. When a contact sits in two campaigns and the ceiling has room for only one, the purpose with the stronger reason goes first: Service notice and Requested contact before Unrequested offer.

A dormant lead list is a reactivation campaign, so in this guide it is an Unrequested offer. Our guide to dormant CRM lead reactivation adds the list checks and retry states for that case.

An example: 100 quote requests

Example, not real data. A Requested contact campaign calls 100 people who filled in a quote request form. The policy is the one above: an attempt cap of 6 over 5 working days, at most 2 attempts a day, and each retry in a different time block from the attempt before.

Attempt 1 places 100 calls and reaches 40 people: 21 Completed, 9 Call me later, 7 Not interested, 2 Wrong number and 1 Do not call. The other 60 calls end in a line result: 41 No answer, 11 Busy and 8 Voicemail.

The 9 callbacks happen at the agreed times, and they are not attempts. In this example, every callback connects at the agreed time. So the retry pool for attempt 2 is exactly the 60 line results. People reached at a later attempt also leave the pool, whatever they said. The calls placed at each attempt are the calls placed at the attempt before, minus the people reached there.

In total, 280 attempts reached 80 of the 100 people. That is a reach rate of 80%: people reached divided by people on the list. It is also 280 / 80 = 3.5 attempts per reached contact: all attempts placed divided by people reached, with callbacks not counted. The other 20 people were not reached after 6 attempts each.

What the attempt yield shows

Retries doubled the reach rate. After attempt 1, 40 of 100 people were reached (40%). After attempt 6, 80 of 100 were reached (80%). The 40 extra people took 180 extra attempts: 60 + 42 + 31 + 25 + 22 = 180.

Later attempts reached fewer people per call. Attempts 2 and 3 reached 29 people with 102 calls (18 + 11 people, 60 + 42 calls), or 28%. Attempts 4 to 6 reached 11 people with 78 calls (6 + 3 + 2 people, 31 + 25 + 22 calls), or 14%. Attempts 5 and 6 alone reached 5 people with 47 calls (3 + 2 people, 25 + 22 calls), or 11%.

Stopping at 3 attempts would have reached 69 of 100 people (69%) with 202 attempts (100 + 60 + 42), and left 31 people not reached.

The cumulative curve hides part of this. The attempt yield falls attempt by attempt: 40%, 30%, 26%, 19%, 12%, 9%. The people still not reached are harder to reach. The cumulative curve, the "Reached so far" column, flattens too, but partly because fewer people are left to call: from attempt 3 to attempt 6, the calls placed fall from 42 to 22. So a flat cumulative curve alone does not show whether late attempts stopped working. The attempt yield does.

The purpose decides what the yield means. For this Requested contact list, attempts 4 to 6 are part of keeping a promise: those 11 people asked for the call. So the cap of 6 holds. If the same curve came from an Unrequested offer, it would argue for a cap of 3: 78 more calls to people not reached in three attempts, for 11 conversations. That is why the cap belongs to the purpose, and why you check it against the attempt yield of the last campaign with the same purpose.

What to measure after each campaign

After every campaign, read these per purpose:

• Reach rate: the share of the list you reached. In the example, 80 of 100, or 80%.
• Attempt yield by attempt: which late attempts still reach people.
• Attempt yield by time block: which parts of the calling day reach people on this kind of list, and so where the next retries should go.
• Attempts per reached contact: how many attempts it took, on average, to reach one person. In the example, 3.5.
• Not interested and Do not call by attempt number: the share of people reached at each attempt who gave one of these answers. If it rises on late attempts, the late attempts cost goodwill.

Then change the cap based on what the yield shows, one purpose at a time. What counts as enough yield depends on the purpose, as the example shows. Keep the last attempts if they still reach enough people for the purpose and the share of Not interested and Do not call does not rise on them. Lower the cap if they reach too few people, or if that share rises. Compare each change with the last campaign that had the same purpose.

Where DRING fits

Reach, DRING's outbound calling product, works a list by rules like these. Reach does not set its own limits. Your team sets them before a campaign runs:

• Calling windows: a start time, an end time and the days a contact may be called, with pauses for quiet hours and holidays.
• Daily and weekly limits: a maximum number of attempts per contact, per day and per week.
• Retry by result: each line result carries its own delay: No answer, Busy, Voicemail, and a call that fails to connect.
• Callbacks: when the agent offers a time and the contact picks one, Reach calls at that exact time, not as a standard retry.
• Do not call: a do-not-call flag or a withdrawn consent removes a contact from every future attempt, in every campaign.
• Best call-time analysis: inside your rules, the next attempt is placed around when the contact has been reachable before.
• Stopping as a result: calling stops when the goal is met, the contact declines or a do-not-call flag is set, and the stop is recorded as the outcome.
• CRM write-back: every attempt and outcome syncs to your CRM, so your attempt yield can come from your own records.

On average, the first attempt reaches 30 to 50% of a list, and retries take cumulative reach to about 85%. These are average figures and vary by sector, country and list quality. They are separate from the example campaign above, which was written for this article.

If you are planning an outbound campaign, request a callback. Have the results by attempt from your last campaign with the same purpose ready. Our team can read the attempt yield with you and set the rules for your next list.