A radio outside broadcast works when a sponsor has covered its cost or the next audience wave shows more people in that market listening, and a crowd on the day proves neither. Most stations count contacts and call that success, which measures effort.
Key points:
- An outside broadcast has two jobs, earning money and building audience, so judge them separately.
- Contacts made measure activity, and on-air replies measure the keenest people.
- Only a survey of the market shows whether listening moved.
- One roadshow will rarely show in a county figure, so test a season of work.
- Hold back a similar area, or you can only say the figure changed.
What is a radio outside broadcast?
A radio outside broadcast takes a show out of the studio to a market, stage, school, stadium or a sponsor's premises, with a live link back to the transmitter. It often comes with a street promotion, in which a team works the surrounding area.
It is a sellable product. Berur FM's 2022 card priced a live link at KSh 80,000 within Nairobi and KSh 120,000 outside it, and KTN Home's 2022 card priced live television transmission at KSh 1,000,000 an hour in Nairobi.
The field side has its own trade. A promotions coordinator runs it, brand ambassadors and volunteers do the walking, and fan club members bring energy and a crowd. Takwimu Intelligence (TI) keeps this work in one place: route lists, promotion areas, promotion scripts, promotion results, contacts made and payments.
What is an outside broadcast for?
It has two jobs, and most stations blur them. One is to earn money, because a brand pays to be present. The other is to build audience, because people who meet the station may start listening.
Separate them before you leave the building. Write down which job this event is doing, and how you will judge it.
If a sponsor pays more than the event costs, the money job is done, whatever happens to the audience. If no sponsor pays, the event has to earn its keep through audience gain, which is a much harder test.
What does it cost?
Add up everything. Here is an invented one-day roadshow.
| Cost | KSh |
|---|---|
| Crew and presenters | 60,000 |
| Transport and fuel | 40,000 |
| Live link and equipment | 50,000 |
| Brand ambassadors, ten at KSh 2,000 | 20,000 |
| Branded materials | 30,000 |
| Lost time in the studio and sales team | 20,000 |
| Total | 220,000 |
Now compare. A sponsor paying KSh 300,000 leaves KSh 80,000, and the audience gain is a bonus. With no sponsor, you have spent KSh 220,000 on a day out and a sales team that was not selling.
Stations often leave the last line out. It belongs there. A roadshow that takes the sales team off the road for a week has cost much more than the fuel.
What counts as evidence that it worked?
There are three levels, and each claims more than the last.
| Level | What you can count | What it proves |
|---|---|---|
| Activity | Events run, contacts made, materials given out | The team did the work |
| Response | Calls, texts, sign-ups, entries | Keen people replied |
| Effect | Reach and share in the next wave | Listening moved in that market |
The trouble starts when a report quotes level one and implies level three. Twelve thousand contacts is a fact about your team. It is silent on whether anybody listened the next morning.
Treat responses with the same care. A study can be announced on air, with replies by USSD, SMS, voice or web through the outlet's own gateway account, and a code per station shows which station brought replies.
Those replies are self-selected, so they never measure reach. Our page on why on-air responses do not measure reach explains why.
Why can't one roadshow show up in a county figure?
The sums are against it. A county holds hundreds of thousands of adults, and an event might meet a few thousand of them. A survey cannot see a change that small.
Here is what a wave can carry, using a 20% figure and comparing the change in one county with the change in another. The margins are simple sampling margins, and weighting makes them a little wider.
| Interviews per county, per wave | Margin on a 20% figure | Smallest net change you can claim |
|---|---|---|
| 400 | 3.9 points | About 8 points |
| 1,000 | 2.5 points | About 5 points |
| 1,600 | 2.0 points | About 4 points |
So plan for a change a wave can see. A season of work in one county, with on-air promotion, outside broadcasts and street teams together, may clear that bar.
One Saturday will not, and it is wrong to claim it did. Our page on sample size for an audience survey covers the arithmetic.
How do you test whether promotions worked?
Hold something back. Pick two counties that look alike on their last wave, run the campaign in one and leave the other, then measure both.
Here is an invented result. In the campaign county your main-station figure goes from 18% to 26%.
In the held-back county it goes from 19% to 20%. The net change is 7 points, which clears the bar at 1,000 interviews a county, and now you can say the campaign moved it.
Without the comparison you can say only that the figure changed. TI's campaign record uses exactly that language: a figure is said to have moved something only where a held-back test earned it, and everywhere else it changed and there was no comparison. It also holds what was recommended, what was decided and how it scored.
Promotion areas need to be real for this to work. A county is the practical unit, because that is where answers are weighted to KNBS census totals. Our guide to weighting an audience survey explains why.
Where should you go?
Go where you can win. Most stations take the roadshow to their strongest town, because the crowd is easy and the presenters are loved there. That spends money on people who already listen.
Use the county table. Campaign tracking on TI compares your station with your main competitor, county by county, on who was listened to most yesterday. The useful target is a county where you are close behind, the signal is sound and your reasons for leaving look fixable.
Check reception first. A roadshow in a county where people cannot hear you the next morning is money thrown away.
What should happen on air around an event?
An event with no air support is a private party. Promote it for a week beforehand, carry it live, and talk about it for a week afterwards.
Give people a reason to tune in the next day. A show name, a time and something that continues tomorrow turn a street contact into a possible listener. A T-shirt alone does not.
Then repeat. Several visits to one area in a season do more than one visit each to twelve towns, and they give a wave something to find. Our page on how to grow a radio audience sets out what must be right before you promote.
How do you sell an outside broadcast to a sponsor?
Sell it as a package, ahead of the date. A sponsor is buying presence, airtime around the event and a live link from their premises or their market.
Put the on-air parts in writing: promos, mentions and live links, each with a count. Then set out what you will report afterwards, which should be what ran and the activity on the day. Promise no audience gain from one event, and our page on sponsorship evaluation shows why.
Sell the season to larger clients. A sponsor who funds six visits to two counties gives you the budget for the test on this page, and gives the client a much stronger story.
Was it worth the money?
Turn the gain into shillings. Here is an invented example. A season of promotion costs KSh 1,500,000 in a county of 500,000 adults, and the net gain is 5 points, or 25,000 adults.
That is KSh 60 for each adult gained. At a target of KSh 200 per thousand, 25,000 more adults supports KSh 5,000 more on a 30-second card price. If half of that sticks after discounts, across 300 spots a month, you gain KSh 750,000 a month, and the season pays back in two months.
Check the claim at renewal. The money arrives only if buyers accept the higher rate, so the sales team needs the new reach note in hand.
How should you pay and manage a street team?
Pay for the work you can check. Brand ambassadors paid per contact will find contacts, and some of them will not be real.
Use route lists and promotion areas, so that a coordinator knows who was where. Give the team a promotion script, then spot-check it. A script that says the show name, the frequency and the time does more than a free T-shirt with no message.
Use fan club members well. They are excellent at bringing energy to an event, and they are the least useful people to ask whether the station is growing, because they already love you. They do not belong in a sample.
What rules apply?
Three sets matter. The Programming Code says broadcast competitions must be run fairly, with rules and any costs to the audience disclosed, and prizes described accurately.
Gaming is stricter. The Code says broadcasters must obtain the relevant licence from the Betting Control and Licensing Board to air gaming campaigns, promotions and activations, so check any draw or game with that in mind.
Personal data needs care as well. Phone numbers collected at a roadshow belong in a consented list with a stated purpose, and our page on data protection for audience research sets out the duties. Keep sign-up lists apart from survey answers.
What do stations get wrong?
The commonest mistake is reporting the crowd. A big crowd at a free concert tells you that people like free concerts.
The others are going where the station is already strong, running one event with no follow-up on air, and never comparing with anywhere. The strong-market habit feels good and teaches nothing.
What should you do on Monday?
Review your last three events.
- Write down what each cost, including staff time.
- Note what the sponsor paid and which job each event was doing.
- Pick the county where you are weakest and could still win.
- Choose a similar county to hold back.
- Plan one season of work there, between two waves.
Read next
- Audience measurement software for Kenyan media: the TI pack page for this work
- How radio and TV stations grow their business with audience data: all the pages in this section
- How to grow a radio audience
- Sponsorship evaluation
- Why on-air responses do not measure 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
A crowd is pleasant and a wave is evidence. TI holds the field work and the audience figures in one place: route lists, promotion results and payments on one side, reach and share by county on the other, with the campaign record showing what moved and what only changed. Replies stay apart from reach.
To see it on your own station or title, book a walk-through.