Skip to content

guide

How a TV station uses audience data to earn more

TV station audience data pays when it decides what a spot can cost, which programmes earn another season, who fronts the news and whether a schedule change worked.

Takwimu editorial team, Editorial team13 min read

TV station audience data earns money when it decides what a spot can cost, which programmes earn another season, who fronts the news and whether a schedule change worked. A television hour costs more to fill than a radio hour, so a wrong decision costs more and takes longer to put right.

Key points:

  • Free-to-air television took almost all TV advertising in Kenya in April to June 2026, and most of its value sits in a few evening hours.
  • A prime-time rate holds when you can show reach and who is in it, band by band.
  • Sponsors buy programmes, so a programme's reach and appreciation decide the renewal.
  • Reasons for leaving tell you whether to change the show, the adverts or the signal.
  • The next wave shows whether a change moved anything, and the campaign record keeps what was tried.

Why does a TV station need its own audience data?

Television is still the largest buy. The Communications Authority of Kenya (CA) recorded KSh 7,939 million of TV advertising in April to June 2026, out of KSh 14,328 million across TV, radio and print. It also found 74% of people watched television in the past seven days and 52% daily.

The competition has changed. The CA says digital platforms attract budgets through targeted reach and measurable engagement, and it warns that broadcasters may be less able to depend mainly on traditional advertising. Nation Media Group's broadcasting revenue fell from KSh 1,360.5 million in 2023 to KSh 1,173.2 million in 2024, across East Africa, and Standard Group's fell from KSh 932.7 million to KSh 735.0 million.

National figures show who leads. They do not tell you why viewers left your 8pm drama in one county, or what a mid-sized advertiser should pay for your 7pm news. Your own wave does.

What does a TV audience wave measure?

It asks people about yesterday. Takwimu Intelligence (TI) splits the television day into five bands: before noon, noon to 5pm, 5pm to 8pm, 8pm to 11pm and after 11pm. It asks which one channel they watched most, and on how many of the last seven days they watched.

The programme questions go deeper. People say which programmes they saw in the last seven days, which they saw most often and what mark out of ten they give it. A programme panel adds what they like, where they catch it and the one thing they would change.

Three facts stay apart. Coverage is who could receive you, reach is who tuned in, and appreciation is what they thought. Streaming logs, on-air responses and advert logs measure other things, and none of them is ever added to a recall figure.

How do you win and keep advertisers?

The decision is which accounts to chase and what evidence to bring. Start from who spends.

In April to June 2026 the CA put media and publishing at KSh 1,588 million on television, banking and finance at KSh 1,144 million and telecommunications at KSh 942 million. Foods and hygiene and beauty care each passed KSh 600 million.

Those buyers already hold national figures. What moves them is evidence they do not have: who your viewers are in the counties they care about, what share of those viewers you hold, and how those people compare with the population. The Audience wave report carries daily television reach, reach at prime time, the channel watched most and cross-tabs by county, age and area type.

Take it to the client before they ask. An extract of two pages beside the rate card is enough, with the method stated so the caveats travel with the number. A buyer who can defend your figure to their own board is a buyer who renews.

Sponsors need more than reach. They want to know the programme is liked by the people watching it, so bring appreciation with the reach figure. Our guide to sponsorship evaluation sets out what a sponsor should ask.

How do you defend the rate card?

The decision is where prices hold and where they should move. Published cards show how steep the curve is. KTN Home's 2022 card priced a 30-second spot at KSh 156,500 from 8:50pm to 9:55pm and KSh 70,500 off prime, more than twice as much for the same 30 seconds.

Discounting eats that premium. Business Daily reported in June 2009 that Synovate estimated TV discounts averaged 40%. That estimate is old, and it shows how long the pressure has run.

Here is an invented comparison of one channel's two evening bands, on one survey.

Band Viewers naming it their main channel 30-second rate Cost per thousand
5pm to 8pm 150,000 adults KSh 60,000 KSh 400
8pm to 11pm 200,000 adults KSh 100,000 KSh 500

The earlier band is the cheaper buy per thousand. So the sales manager has a case for raising it, and a reason to refuse a discount on the later one. Our page on TV advertising rates in Kenya shows how rates are set.

Definitions need care. The Programming Code defines TV primetime as 6pm to 10pm, the CA report names 7pm to 10pm as the strongest advertising window, GeoPoll's first-quarter 2026 report uses 7pm to 9:30pm, and TI's prime-time band is 8pm to 11pm. Name your band every time you quote a figure.

Sell what was measured. A band figure counts people reached in that band, so it stays a reach figure. It carries a margin of error on a probability sample, and on a quota sample it describes respondents and is never a rating.

How do you price a sponsorship?

Start from the airtime it contains. KTN Home's 2022 card listed property sponsorship at KSh 850,000 a week and programme sponsorship at KSh 750,000, with a bonus value of seven minutes a week split 40:60 between prime and non-prime.

Here is an invented example of that arithmetic. Seven minutes is 14 units of 30 seconds, so 5.6 prime units and 8.4 non-prime units. At KSh 100,000 and KSh 45,000 a unit the airtime is worth KSh 938,000 at card.

The sponsor then asks what the association adds. Appreciation is your answer. A programme with a strong mark out of ten and few people switching away is a better place for a brand than a larger programme people resent, so show both figures at the renewal.

How do you fix the schedule?

The decision is what gets another season, what moves and what ends. Television makes that decision expensive. Here is an invented example of a local drama against an acquired series for one 26-episode slot.

Local drama Acquired series
Cost per episode KSh 300,000 KSh 60,000
Cost for 26 episodes KSh 7.8 million KSh 1.56 million

The drama costs KSh 6.24 million more. To pay for itself it must bring in KSh 240,000 more per episode in sponsorship and spots. Reach in the band, appreciation among viewers and the share who switched away tell you whether that is plausible before you sign.

Read the band pair first. If the 5pm to 8pm band is strong and 8pm to 11pm collapses, people are leaving after the news, and the show that follows it is your problem. Our page on prime time TV in Kenya covers how the evening is built.

TI also warns when most people cannot place a single programme. Fewer than a quarter naming any programme means your programme figures rest on too few people. Promotion comes before any change to the grid.

How do you stop losing viewers?

The decision is what to fix first. People who switched away part-way through a programme say what made them go. Those reasons are coded into nine themes, and cross-tabs show whether a complaint belongs to one county or to everyone.

Three readings do most of the work. Left because of advertising says your breaks are too heavy for that programme. The Programming Code says adverts should not exceed ten minutes in any 30 minutes of broadcasting, so the legal ceiling is far above what most audiences will sit through.

Left because of the signal says check distribution before you touch the content. A signal fault tends to sit in one or two counties, while an editorial fault spreads. TI's service level gives a sharp fall in one market 48 hours, because checking costs little and rebuilding a schedule around a fault you never found costs a year.

Moved for something specific points to a rival's event or sport. That one is harder to fix and easier to plan for. Our page on why audiences leave explains how the reasons are built.

How do you back the right anchors and presenters?

The decision is who gets the contract and the 7pm chair. Talent costs more on television, and a sponsor often buys the face as much as the slot.

Station-wide numbers hide this. An outlet can be losing audience to one presenter and holding it with another, and a single figure averages the two. Read left because of the presenter beside held by the presenters, then read the Programme performance report show by show.

Bases are small at programme level. Only people who could place a programme are counted, so a reading can rest on 150 interviews with a margin of about 8 points. Use one wave to open the conversation and two to support a contract.

How do you prove a change worked?

The decision is whether to keep it. Measure again with the same questions in the same population, then read the change against its margin. Two waves that each carry 3 points leave about 4 points on the difference.

A move past that margin shows the figure changed. It shows your change caused it only where something was held back, such as a promotion run in one county and withheld in a similar one. Everywhere else the record says it changed, and there was no comparison.

The campaign record keeps what was recommended, what was decided and how it scored. That protects the station from repeating expensive ideas when the people who remember them have left.

What should a county or vernacular channel do?

Own a place. The CA counted 241 TV stations operating at 31 December 2024, and most of them will never top a national table. A county channel that shows it leads in one county, in one language, has something a national seller cannot offer.

The CA's August 2026 report names regional and vernacular broadcasting as an opportunity, because regional differences show continuing demand for local, language-specific content. So measure your own county, and take that evidence to the county's supermarkets, colleges, SACCOs and hospitals.

Read national tables with care. A national sample holds few people from one county, so a small channel can look absent when it is simply unmeasured there. Our page on vernacular TV stations in Kenya sets out what is known.

How should you treat YouTube and social numbers?

Sell them as a separate line. The CA says broadcasters can extend reach by repurposing programmes for websites, YouTube and social media, and many do. Platform views are counted by the platform on its own definition.

A view is a different thing from a person who watched yesterday. Keep clip views in their own column and never add them to a recall figure. Our page on pay TV and streaming in Kenya covers how viewing is shifting.

Plan the income modestly. In a Media Council of Kenya survey of media professionals in October and November 2025, 52% said digital platforms brought under 10% of income.

Is a survey wave worth the money?

Set it against the size of the deal. KTN Home's 2022 card listed programme sponsorship at KSh 750,000 a week, so a 13-week sponsorship at that price lists at KSh 9.75 million before VAT. A wave that helps keep one such client has done its job.

Check the cost on our pricing page. Then run a wave when a decision of that size is on the table, and leave it when none is.

How does this sit beside national ratings?

They do different jobs. The national picture comes from sources such as GeoPoll, Ipsos and the CA, and agencies plan against it. Your wave gives you depth in your own market and the reasons behind the number.

Keep them side by side and never merge them. Our page on TV ratings in Kenya explains how to read the national figures.

What do TV stations get wrong?

Four errors recur. The commonest is quoting daily television reach as if it were one channel's audience, when it belongs to the medium.

The second is treating a strong 7pm bulletin as proof of the whole evening. The third is renewing a show out of affection, with nobody having checked who watches it or why they leave.

The fourth is arriving at a sponsorship renewal with reach alone. The sponsor has seen that number. Bring what the audience thought of the programme.

What should you do on Monday?

Start with the money.

  1. List your ten largest advertisers and when each one renews.
  2. Mark the three programmes whose future will be decided in the next six months.
  3. Write what each decision is worth, using your own sponsorship and production costs.
  4. Ask your engineers for distribution faults by county in the past quarter.
  5. Agree which counties matter to those advertisers, then plan one wave around them.

Where TI fits

Television spends too much per hour to run on instinct. TI measures audiences by asking people, weights answers to KNBS census totals at county level and shows each figure's margin of error, then reports reach by band, channel share, appreciation and reasons for leaving. The channel owns its data. Its broadcast monitoring keeps a separate record of what went out on air, so a figure can be read against what ran that week.

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, Quarter 4 2025/26 · Communications Authority of Kenya, August 2026Seven-day and daily media use, radio-only use, advertising spend by medium and sector, the evening advertising window, challenges and opportunities for broadcasters. Opened 5 November 2026.

  2. 2

    Nation Media Group 2024 Annual Report and Financial Statements · Nation Media Group PLC, June 2025Notes 5 and 6 on segment results and revenue by product line, the Government of Kenya as a customer, credit periods. Opened 5 November 2026.

  3. 3

    The Standard Group PLC financial statements for the year ended 31 December 2024 · The Standard Group PLC, June 2025Directors' report on how each platform earns, notes 7 and 8 on segment results and revenue. Opened 5 November 2026.

  4. 4

    KTN Home advertising rate card 2022 · Standard Group, 2022Spot, sponsorship, branded feature and live transmission prices, and terms. Opened 5 November 2026.

  5. 5

    High entry barrier denies TV advertising spend · Business Daily, 21 June 2009Discounting estimates for TV and radio. Opened 5 November 2026.

  6. 6

    The Programming Code for Broadcasting Services in Kenya, 4th Edition · Communications Authority of Kenya, March 2024Primetime definitions, clause 11 on advertisements and clause 12.4 on broadcast competitions. Opened 5 November 2026.

  7. 7

    The Top TV and Radio Stations in Kenya, Q1 2026 · GeoPoll, 21 April 2026January to March 2026 reach and daypart shares. Opened 5 November 2026.

  8. 8

    Second Quarter Broadcasting Services Report, FY 2024/25 · Communications Authority of Kenya, 2025Licensed and operational broadcasters at 31 December 2024. Opened 5 November 2026.

  9. 9

    Navigating the Digital Reality: Monetisation Challenges and Opportunities for Kenyan Media in the Digital Economy · Media Council of Kenya, January 2026Survey of 17 October to 23 November 2025: digital share of income, digital revenue strategy, foreword on advertising. Opened 5 November 2026.

More from the library