Skip to content

guide

The audience wave report on TI

What an audience wave report contains, section by section, and how an advertiser, a programme controller and a regulator should each read it.

Takwimu editorial team, Editorial team10 min read

An audience wave report sets out what one round of fieldwork found, in a form that an advertiser, a programme controller and a regulator can each read without being misled. On Takwimu Intelligence (TI) it opens with its own limits, then runs from the sample through reach, share, dayparts, programmes and reasons.

Key points:

  • The limits sit on the first page, where a reader meets them before any figure.
  • Every figure comes from asking people what they did yesterday.
  • Reach is a share of whoever answered, and it speaks for a population only where the sample was drawn so that it can.
  • Station share is a share of listeners who named a main station, and it says nothing about listening time.
  • The report ends with reasons, coded voices and the methodology, so the numbers come with an explanation.

What does an audience wave report contain?

The report follows one fixed order. That order is a large part of its value, because readers learn where to look.

  1. What this report measures, and what it does not.
  2. Interviews achieved, and how the people who answered compare with the population.
  3. Reach by medium: daily and weekly radio, daily television, daily newspaper, any medium yesterday, and radio and not television.
  4. The station listened to most yesterday.
  5. Radio reach at breakfast, drive time and in the evening, with television reach at prime time.
  6. How people listened.
  7. Programme figures: could place at least one programme, the appreciation index and switched away mid-programme.
  8. Why they left, left because of advertising, left because of the signal, held by the presenters and held by habit alone.
  9. Cross-tabulations, what the audience said in their own words, and the methodology.

Why do the limits come first?

A footnote does not travel. Whoever forwards a figure rarely forwards the caveat, so TI puts it where it cannot be skipped.

The opening section says every figure comes from asking people what they did yesterday. Day-after recall is used because how often somebody says they usually listen is not behaviour, and that gap is the oldest known problem in this field.

It then says what reach is. Reach is a share of whoever answered. It describes a population only where the sample was drawn so that it can, and the sampling design recorded on the study says in writing which applies. Where the design is a non-probability one, no figure in the report is a rating and none may be sold as one.

How does the report show who answered?

Two sections carry this. Interviews achieved gives the count. How the people who answered compare with the population sets the sample beside the census.

Read this section before you read reach. If young men are under-represented, the weighting will have worked hard, and figures for that group need care.

One warning helps here. Where television reached more people yesterday than radio, the report raises an information note. It may be a real feature of the market, or a sign that the sample leans towards people who are easier to reach online. This section is where you settle which.

Which reach figures does the report carry?

Each figure has its own base, so read them in this order.

Figure What it reads How to read it
Daily radio reach Listened to any radio yesterday The size of the pool every station competes inside
Weekly radio reach Listened on at least one of the last seven days Always larger than daily reach
Daily television reach Watched any television yesterday Share of those who answered
Daily newspaper reach Read a newspaper yesterday, on paper or on a phone Reading on a phone counts as reading
Reached by any medium yesterday Used radio, television or a newspaper Base is people who answered all three
Reached by radio and not by television Listened and did not watch The part a television campaign cannot buy

The gap between weekly and daily shows how occasional an audience is. As a public benchmark, the Communications Authority of Kenya's April 2026 report puts radio's seven-day use between 73% and 75% over the four quarters to Q3 2025/26, and daily use at around 51% in Q3 2025/26. Those are its figures from its own survey. Your market will differ, and that difference is the finding. See daily reach and weekly reach.

Why does a daily radio figure not belong to one station?

Daily radio reach belongs to the medium. It says how many people listened to any radio, and every station competes inside that pool. Reporting it as a station's own audience is the most damaging error a sales team can make with this report.

The station's position comes from the section after it. The station listened to most yesterday shows every outlet on the prompted list, and that is where a station finds itself against its competitors.

How should station share be read?

Station share is the share of yesterday's radio listeners who listened to the client's station most. It is a share of people, and it says nothing about hours.

That matters when someone sets it beside a figure from elsewhere. In the UK, RAJAR calculates share from hours, as a station's hours divided by all radio hours in its area. The same word can describe two different measurements, so name the base whenever you quote one. Our guide to rating vs share goes further.

Read share against the main competitor. A station can hold share while the pool shrinks, or lose share while its own audience grows. You need both figures.

How do dayparts read in the report?

Breakfast comes first, because in almost every radio market it is the largest daypart and the one advertising is priced from. Our page on breakfast radio covers what that means for a schedule. Where breakfast is smaller than the afternoon, the study raises a warning. Either something is wrong with the breakfast show, or the market's day starts differently from the assumption the schedule was built on.

Drive time is the second peak. It moves with how people travel, so a change there may have more to do with matatu use than with the schedule. The evening is where radio competes with television directly, which is why it sits beside television reach at prime time.

The CA's April 2026 report puts the strongest television advertising window at 7pm to 10pm, when audience size and advertising activity peak together. Read your own bands against that.

Daypart reach and average quarter-hour are different numbers

Daypart reach counts people reached at some point in a band. Average quarter-hour (AQH) counts people listening in an average quarter-hour inside it. Advertisers price those differently.

AQH needs finer data than daypart recall collects: half-hour recall, a metered panel, or a diary such as RAJAR's one-week diary, where people log each quarter hour as they go. The wave report carries daypart reach and states that it carries no AQH, because a number computed from daypart recall and given that name would be a different measurement wearing a familiar label.

Quote the figure you have. Buyers respect that.

How does the report show the way people listened?

One question gives the main route yesterday: a radio set at home, a radio in a vehicle or matatu, FM on a mobile phone, streaming over the internet, or through a television set.

Two indicators follow from it. Listened mainly over the internet tells you how much the data bill matters beside the transmitter. Listened mainly in a vehicle tells you how far your peaks follow the commute.

Neither is good or bad in itself. They are facts to plan around.

What does the programme section add?

Visibility comes first. Could place at least one programme is the share who named a listed programme as the one they hear most. Where it is low, every programme figure rests on a thin base.

The appreciation index is the average mark out of ten for the programme somebody named, multiplied by ten. Switched away mid-programme is recalled over a month, so it shows how common the behaviour is and leaves open how often it happened.

The programme performance report takes these show by show, and our page on the appreciation index covers what the score can and cannot carry.

What does the report say about why people leave and stay?

This is the half of an audience figure that says what to do. The report sets out why people last switched away, then pulls out advertising and signal, because those two belong to sales and engineering.

Two figures describe what holds the audience. Held by the presenters is real audience, and it can leave with one resignation. Held by habit alone is the most fragile audience there is, because nothing has to go wrong for it to leave. Our page on why listeners switch stations takes each cause in turn.

How do cross-tabs and open answers help?

Cross-tabs show where a figure lives. A main station by area type shows whether you are strong in towns and weak in rural areas, and reasons for leaving by area type shows whether one complaint belongs to one kind of place.

Open answers go through two coding frames. A respondent's words are coded, so "matangazo mengi" and "too many ads" count as one complaint. Left apart, they would look like two findings at half the size.

The methodology section closes the report. It states how people were selected in plain words.

Who reads which part of the report?

Three readers want three things.

Reader Starts at Wants to know
Advertiser or media buyer Limits, sample, reach, station share Is this a figure I can plan on?
Programme controller Dayparts, programme figures, reasons What do I change, and when?
Regulator Limits, methodology How was this produced?

Send each reader to their page. For a shorter document, the reach note for an advertiser carries the same caveats with fewer figures.

What goes wrong when a wave report leaves the building?

Trouble starts with extracts. These five are the ones to stop.

  • A slide quotes daily radio reach as the station's audience.
  • A figure from a quota sample goes out as a rating.
  • A programme score is quoted with no base.
  • Daypart reach is relabelled as AQH for a media plan.
  • A survey figure is added to streaming sessions to make a bigger number.

Our guide on how to report an audience wave sets out house rules that stop each of these.

Where TI fits

A report earns trust through its limits. TI builds the wave report from the study's own sampling design, weighting and coded answers, so each figure reaches the reader with its base and the way it was made.

To see it on your own station or title, book a walk-through.

Takwimu Intelligence (TI) is audience measurement software for Kenyan media.

Sources

  1. 1

    Audience Measurement and Industry Trends Report, abridged version, April 2026 · Communications Authority of Kenya, April 2026Seven-day and daily media use to Q3 2025/26, radio genre and language, advertising window, industry challenges. Opened 5 November 2026.

  2. 2

    The RAJAR measures of listenership · RAJARDefinitions of weekly reach, weekly hours and share. Opened 5 November 2026.

  3. 3

    How the survey is conducted · RAJARThe one-week diary, quarter-hour entries and the ban on self-selection. Opened 5 November 2026.

More from the library