Media planning is deciding who an advert must reach, how often, where and through which outlets, before any money is committed. In Kenya the work has one extra step: planning by county and language, because a national average hides most of what decides a campaign.
Key points:
- Start with the people and places, then choose media.
- Reach and frequency are the core numbers, and gross rating points multiply them.
- Never add station reach figures together, because shared listeners get counted twice.
- Radio and TV each reach about three in four Kenyans in a week, by the CA's April to June 2026 figures.
- Plan the measurement before the booking, or the plan cannot be judged.
What is media planning?
Media planning is the bridge between a marketing goal and a schedule. The planner turns "make farmers in Western Kenya aware of this seed" into stations, dayparts, weeks and spots.
Media buying follows. Planning decides what is wanted, and media buying in Kenya gets it at a price. In a small team one person does both, which is why the two get confused.
What is the difference between media planning and media buying?
Planning is the decision, and buying is the deal. A plan says "reach 65 per cent of farmers in four counties five times". A buy says "20 breakfast spots on this station at this net price".
The two jobs need each other. A plan with no regard for price cannot be bought, and a buy with no plan is shopping.
What is the media planning process?
Most good plans follow seven steps:
- Write the goal as a change in people, with a number and a date.
- Define the audience by age, gender, language and county.
- Set the reach and frequency needed.
- Choose the mix of radio, TV, print, digital and outdoor.
- Build the schedule by week and daypart.
- Cost it against current rates.
- Agree how results will be measured before the first spot runs.
Teams usually skip step 7. It costs nothing to write down and everything to reconstruct afterwards.
Who are you trying to reach?
Define the audience in the terms your data can see. The CA's report for April to June 2026 breaks use down by age, gender, setting and living standard measure, and GeoPoll's dashboard filters by demographics and location.
Language narrows the choice fast. The CA finds Swahili the main language of radio listening, with over 60 per cent of male listeners and around mid-50 per cent of female listeners on Swahili stations, and vernacular stations second, stronger in rural areas. Our guide to vernacular radio stations goes further.
Age splits the digital plan. The CA says internet use tops 70 per cent among people aged 15 to 24 and falls below 40 per cent at 45 and above.
What is reach and frequency in media planning?
Reach is the share of your audience who see or hear the campaign at least once. Frequency is the average number of times those people are exposed.
Here is an invented example. If 60 per cent of your target hears an advert an average of four times, reach is 60, frequency is 4, and the plan delivers 240 gross rating points. Our page on reach and frequency has more examples, and gross rating points explains the formula.
The same 240 points could be 80 per cent at three times, or 30 per cent at eight. Those are very different campaigns.
How much reach do you need?
Fit it to the job. A new product launch needs wide reach, because most buyers have never heard of it. A reminder for a known brand can run lighter and longer.
Ask what the campaign must change, then ask how many people must hear it to change that. A planner who starts from the budget ends up describing the budget.
Why can't you add station reach figures together?
Because audiences overlap. Take an invented county where Station A reaches 40 per cent and Station B reaches 30 per cent in a week.
If 12 per cent listen to both, the combined reach is 58 per cent. Adding gives 70, which overstates it by 12 points.
The CA finds many people use several media, so overlap runs across radio, TV and the internet too. Net reach needs one survey asking the same people about every outlet. Separate figures from separate sources cannot produce it.
This is the most expensive error in Kenyan planning. It makes a schedule look bigger than it is, and nobody finds out.
How do you choose the media mix?
Start with where the audience is. The CA's April to June 2026 figures for use in the last seven days are 75 per cent for radio, 73 to 74 per cent for TV, 57 per cent for the internet, 17 per cent for newspapers and 6 per cent for magazines.
Then consider where they are when they use it. The CA finds 87 per cent of radio consumption and 88 per cent of TV consumption happens at home. Only 2 per cent of radio listening was in a bus, taxi or matatu.
That last figure challenges a common belief about matatu radio. Treat it as one source's finding, and test it in your own county.
Which daypart should you plan for?
Follow the audience, then check the price. GeoPoll's January to March 2026 data found the leading radio stations peaked at 7am, and it defined TV prime time as 7pm to 9:30pm. The CA's report points planners to morning and early evening for a better balance of reach and cost.
Different sources draw these blocks differently. Check the hours before you compare.
What is a media planning strategy?
Strategy here means how the weight is spread over time. Three shapes cover most campaigns.
Continuity runs at a steady level, which suits products bought every week. Flighting runs in bursts with gaps, which stretches a small budget. Pulsing keeps a low steady base and adds bursts at key moments, such as planting season or back-to-school.
Pick the shape from how people buy. A fertiliser brand has a season, and a soap brand does not.
What is share of voice in media planning?
Share of voice is your brand's share of all advertising in a category. Our page on share of voice in media gives the calculation.
Use it to set a floor. If a rival outspends you four to one on radio in your county, a thin schedule will struggle to be noticed.
When should a campaign run?
Timing changes the price. The CA reports that TV and radio spending strengthens in the second half of each year and falls at the start of the next, so January to June is a softer market for buyers.
Election years shift this. Reelanalytics expects political advertising and news consumption to compete for premium space ahead of the 2027 vote. It reported that 30 to 35 per cent of large billboard stock stayed available through much of the first half of 2026, and urged brands to secure locations early.
What does a media plan look like?
Here is an invented plan for a seed company with KSh 3,000,000 to spend in four counties over eight weeks:
| Item | Plan |
|---|---|
| Goal | 50 per cent of farmers aged 25 to 54 aware of the new variety by the end of August |
| Reach and frequency | 65 per cent at five times, or 325 gross rating points |
| Main medium | Two vernacular radio stations, breakfast and early evening |
| Support | One county-level mention package, with a presenter who farms |
| Weeks | Three weeks on, one off, three on, one off |
| Check | Day-after recall survey in each county, before and after |
Each line can be judged afterwards. The goal is a number, the audience is named, and the survey is already in the budget.
What are the main media planning tools?
A spreadsheet still does most of the work. Add three inputs: current rate cards, audience data for your target, and the costs of research.
Commercial tools exist. GeoPoll, for example, describes a media planning tool that lets advertisers find top-rated slots and export to Excel, plus a post-campaign evaluation tool. Reports from the CA are free and carry national trends, while station-level figures usually come from paid sources or a survey of your own.
Our page on running an audience survey covers that last route.
What do planners get wrong?
They plan on national averages for county campaigns. A station that reaches 20 per cent nationally can reach 60 per cent in one county and nothing in another.
They plan on one month's ranking. Rankings differ by source and by method, so ask who measured, when and how. And they plan with no way to check, then call the campaign a success because the spots aired.
How do you know if the plan worked?
Separate three questions: did the adverts run as booked, were the planned people reached, and did anything change in what they know or feel?
Each one has its own evidence. Our page on how a station proves its reach to an advertiser separates the first two, and a brand lift study deals with the third.
Takwimu Intelligence (TI) holds that a plan without a measure is a guess with a schedule attached. Write the survey into the plan on day one.
Read next
- Audience measurement software for Kenyan media: the TI pack page for this work
- Media advertising in Kenya: planning, buying and proving reach: all the pages in this section
- Media buying in Kenya
- Reach and frequency
- Dayparts in radio and TV
- 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
Net reach needs the same people asked about every outlet. TI runs one cross-media day-after recall questionnaire across radio, TV and news, so reach, overlap and reached by radio and not TV come from a single weighted sample at county level, each with its margin of error.
To see it on your own station or title, book a walk-through.