Day after recall is a survey method that asks a sample of people what they listened to, watched or read yesterday, block by block through the day, which gives a station its daily audience from memories no more than a day old.
Key points:
- Day after recall measures remembered use on one day. It gives daily reach, share and daypart patterns.
- Spread interviews across all seven days, because yesterday on a Monday call is a Sunday.
- Weekly reach needs its own question. One yesterday cannot show a week.
- Self-reports can run high yet still follow the trend, so read recall for direction and comparison.
- Compare a recall figure only with one from the same wording and mode.
What is day after recall?
Day after recall covers one day only: the one before the interview. People also write it as day-after recall or DAR, and older European papers call it 24-hour recall.
Ipsos describes its telephone day-after recall as covering the previous day's listening, with stations, platforms and location by quarter-hour. That is the method in its full form. Most station surveys use wider time blocks.
How does a day after recall interview work?
The interview walks a person through yesterday. It takes this order:
- Name the day: "Yesterday was Tuesday."
- Take the person through the day in blocks, from waking up to going to bed.
- Ask whether they listened to any radio in each block, then which station.
- Ask where they were and how they listened: at home, in a vehicle, on a phone, over the internet.
- Repeat for television, then for newspapers and news online.
- Finish with age, sex, county, education and work.
Naming the day matters more than it looks. A person told "yesterday" on a Monday may answer for a normal working day, and Sunday's schedule is another station altogether.
Why do stations use day after recall?
Stations use it because it is short, cheap to repeat and needs no reading or writing from the respondent. It also takes radio, TV and news in one interview, so one person's whole media day sits in one record.
It has a long history. A 1995 ESOMAR paper on Médiamétrie's work counted nine European countries using 24-hour recall and five using diaries.
Kenya's published measurement also asks people at short intervals. The tracking survey KARF supplied to the Communications Authority in 2019 set a daily target of 3,000 people aged 15 and over, interviewed by telephone in 30-minute blocks. GeoPoll says it collects TV and radio data daily through mobile surveys, and reports TV in 30-minute blocks and radio in 2-hour blocks.
What does it measure well?
Recall is strong on a concrete, recent act. Daily reach, station share, reach by time of day and the overlap between radio and TV all come from it cleanly.
There is evidence for this outside media. When Pew Research Center compared phone and web answers on 60 questions, it found an average gap of 5.5 points, yet activities "yesterday" showed little or no difference. A plain question about one recent day travels well.
What does day after recall miss?
It misses what people do not notice. A radio in a shop, a matatu with the volume down, five minutes in a neighbour's kitchen: much of this never reaches memory as "listening".
Nielsen says 45% of the listening occasions its meters record run under five minutes. Those are US figures from a passive system, and they show how much short listening exists for a survey to lose.
Other losses are predictable. People mix up stations with similar names and give credit to the brand they know. Heavy listeners report habit in place of what happened that day.
Do people overstate what they watched?
One well-known study tested this on US television news. Markus Prior compared self-reports with Nielsen ratings in 2009 and found that surveys overstated exposure by about three times on average and by up to eight times in some demographic groups.
The same paper found that survey estimates followed trends in Nielsen ratings relatively well. That is the fair reading of recall. It is weaker as a head count and stronger as a guide to direction and comparison.
Television news in the US is a long way from a Kenyan breakfast show. A question about one named day, in blocks, also gives memory more to hold on to. Even so, do not sell a recall figure as a count of heads.
Day after recall or a diary?
The two methods give different totals for the same behaviour. The 1995 paper reported that diary countries tended to show higher radio figures than recall countries, and that average listening per head under RAJAR rose by 50% when the UK moved from 24-hour recall to a seven-day diary.
People did not listen 50% more that year. The instrument changed. Our page on the radio diary survey covers what a diary adds, and how radio listenership is measured sets all the methods side by side.
| Day after recall | Seven-day diary | |
|---|---|---|
| Covers | One day | One week per person |
| Effort for respondent | One short interview | Daily entries |
| Reading and writing needed | None | Yes, unless assisted |
| Gives weekly habit | Only through a separate question | Directly |
| Cost to repeat | Lower | Higher |
How many interviews do you need on each day?
Plan by day of the week. Here is an invented example with round numbers: 1,400 interviews over a week, 200 for each day of the week.
If daily reach comes out at 30%, the whole sample gives about ±2.4 points. One single day of 200 gives about ±6.4 points, before weighting. So report the week's average and hold back from reading one day against another.
Interview on the day after the day you need. Sunday's listening takes Monday interviews, and Saturday's takes Sunday interviews, which is the shift many field teams try to skip. More on this is in sample size for a media survey.
Can day after recall give weekly reach?
One yesterday cannot show a week. A person who listens twice a week has a good chance of being missed on any one day, and a person who appears today may appear again tomorrow.
Ask a separate question: "In the last seven days, on how many days did you listen to this station?" Multiplying daily reach by seven is a common error, and at 20% daily reach it would give 140%.
The national figures show the gap. The Communications Authority's report for Q4 of 2025/26 has radio at 75% over seven days and 49% daily. Read more on daily reach and weekly reach.
What is the day after recall test in advertising?
The same words name a separate test. In advertising research, a day-after recall test checks whether people remember one advert the day after it ran.
Burke's Day-After-Recall was read as a measure of an ad's ability to "break through" into the consumer's mind. That is a question about one piece of creative. It says nothing about how many people a station reached.
If an advertiser wants memory of a campaign, that is a brand lift study. Keep it apart from your reach figures.
How should a prompted list work in recall?
Most recall surveys read or show a list of outlets, because memory for names is weak. The list then shapes the answer.
Ask with no list first if you want to see which names people hold in mind. Then prompt, and rotate the order. The details are on our page about prompted recall.
What do stations get wrong?
These are the errors that cost most:
- Interviewing on weekdays only, which removes the weekend from the audience.
- Asking "which station do you usually listen to?" and reporting the answer as yesterday's reach.
- Changing the time blocks between waves, which breaks the trend.
- Adding streaming figures to a recall figure.
- Reading a two-point move as growth when the margin is three.
Read next
- Audience measurement software for Kenyan media: the TI pack page for this work
- How to run an audience survey in Kenya: all the pages in this section
- Radio diary surveys
- Prompted and open recall
- Daily reach and weekly reach
- Takwimu Intelligence (TI) home page
- Audience measurement in Kenya: the guide to this library
- The TI platform and how we check our numbers
Where TI fits
Day after recall is the sensible default for a Kenyan outlet because it can be repeated, wave after wave, in the same words. Takwimu Intelligence (TI) asks about radio, TV and news yesterday with a prompted outlet list and time blocks, and reports station share, daily reach and reach by daypart with a margin of error on each.
To see it on your own station or title, book a walk-through.