Skip to content

guide

The appreciation index: measuring what audiences think of a programme

An appreciation index turns marks out of ten into a score for one programme. Here is how to calculate it and how far it can be compared.

Takwimu editorial team, Editorial team8 min read

An appreciation index is a score out of 100 that says how much an audience liked a programme, usually built from marks out of ten. On Takwimu Intelligence (TI) it is the average mark out of ten given to the programme somebody named, multiplied by ten.

Key points:

  • Reach says how many people came, and appreciation says what they thought. They stay separate.
  • TI's index comes from one question in the survey: a mark out of ten for the programme the person named.
  • Because people mark a show they chose, it rates a favourite.
  • It shares a name with the panel-based index published by the BBC, and it must not be compared with it.
  • An average mark under six out of ten raises a warning, because the sample already leans in the show's favour.

What is an appreciation index?

An appreciation index turns opinion into a single number for one programme. Did the people who came value what they got? Reach cannot say.

A show can draw a crowd and leave it cold. Another can hold a small audience that would be lost without it. Both facts matter to a programme controller, and neither can be read from a reach figure.

Stations often run on reach alone, because reach is what sells. That leaves them blind to a programme losing goodwill while its numbers still look fine.

Where does the appreciation index come from?

One long-running version is British. In July 2011 the BBC described its Appreciation Index as a score out of 100 given to its services by an independently run panel of 20,000 people. It reported that for January to March 2011 all its television channels scored more than 80, and all its major radio networks scored 77 or above.

The Sunday Times reported in May 2021 that the BBC's audience research unit uses a daily online survey of 20,000 people who score the programmes they watch out of ten. It also reported that such figures are rarely published, unlike viewing figures.

Keep one thing in view. That is a standing panel, run daily, on one broadcaster's programmes.

How is the TI appreciation index calculated?

The questionnaire asks people which programme they hear or watch most often, then asks: "Give the programme you named a mark out of ten." The scale runs from very poor to very good. Anyone who chose "None of these" leaves it blank.

The indicator is one line:

mean(q_prog_ai) * 10

An invented example: four people give marks of 8, 7, 9 and 6. The average is 7.5, so the index is 75.

The headline figure pools every programme people named. The figure a controller needs is the breakdown by programme heard most often, which the indicator also carries. It can be cut by sex as well.

What does a score of 70 mean?

A score means little alone. Read it against your own past waves and your other programmes.

The Sunday Times reported BBC guidance for staff that treats 90 or above as exceptional, 85 or over as excellent, 60 or under as poor and less than 55 as very poor. Those bands belong to that panel. They do not transfer to a survey of a different kind.

TI declares one line for its own index: an average mark under six out of ten raises a warning. Everything above that is a matter of trend and comparison inside your own study.

Why can't it be compared with a panel-based index?

The name is shared. The measurement is different.

The pack says so directly: this is an appreciation score computed from the survey's own respondents, and it must not be compared with the panel-based Appreciation Index published by broadcasters elsewhere. The panel, the question and the moment of asking all differ.

TI appreciation index A panel-based index
Who answers Respondents in one wave A standing panel
What they mark The programme they named as heard most often Programmes they watch
When At the interview A daily survey
Compare with Your own programmes and past waves The same panel's scores

If a sales deck says "our show scores above a BBC programme", take the line out.

Why is a low score a serious signal?

This is a rating of a favourite. People chose the programme themselves, so the group marking it leans in its favour.

A mark under six out of ten from that group is therefore no soft signal. It says that even the people who picked the show are lukewarm. The study raises a medium warning, under the label "The programme people name is rated below 60".

A high score needs the same care in reverse. It may only mean that the few who can place the show like it. So read it beside how many can place the programme.

Why must reach and appreciation stay separate?

They answer different questions, so a blend would hide both. This is the core position of the Media Audience pack: a station needs who could receive it, who tuned in and what they thought, as three separate facts.

Read them together in a grid.

High appreciation Low appreciation
High reach Protect it, and find out what holds people Audience may be there from habit, so find out why before it goes
Low reach A good show too few people find: check the slot and the promotion Review the programme

The grid gives four different actions. A single score would give one. The same logic applies to audience health, which stays as four separate readings.

How big a gap between two programmes matters?

Less than most meetings assume can be noise. An index is an average, so it has a margin like any other figure.

An invented example, on a simple random sample at 95% confidence. Suppose marks spread by about 2 points around their average. With 100 people marking a show, the average carries about 0.39 of a mark either way, which is about 3.9 index points. With 400 people it is about 2 index points.

Comparing two shows needs more. With 100 people behind each, the gap has to pass about 5.5 index points before it means much. A show on 74 is level with a show on 71.

How does the programme panel differ?

The programme panel asks for a mark out of ten too, and it adds one more number: how likely the person is to tell a friend to listen or watch. Then it asks which part they like most, where they usually catch it, and what one thing they would change.

Everybody answering has already been selected for having heard the programme, so a percentage from it describes that group. It is a tool for diagnosis, and a focus group can take the findings further.

Keep the two apart in your reports. A wave index and a panel mark are different numbers.

What moves the score?

The reasons sit in the same study. Which part people like most points to what to protect, whether that is the presenter, the music, the interviews, the callers, the news or the competitions.

The one thing they would change points to what to fix. Read those coded answers alongside why listeners switch stations, and test one change at a time. Change five things at once and you will learn nothing from the next wave.

How should appreciation be tracked across waves?

Hold the method still. Use the same line-up wording, the same question and the same sampling design from wave to wave. A show renamed between waves breaks its own trend.

Note what changed on air between readings, such as a new presenter or a new slot. Then use day-after recall figures beside the index, so you can see whether a better mark came with a bigger or a smaller audience.

What mistakes do stations make with appreciation scores?

Most come from treating the score as more than it is.

  • Comparing it with a published BBC score.
  • Adding it to a reach figure to make a single rating.
  • Quoting it with no base.
  • Reading the pooled figure as if it described one programme.
  • Reporting a gap of two or three points as a win.
  • Mixing panel marks into the survey figure.

Each of these is easy to avoid. Print the base, name the programme and keep the score in its own column.

Where TI fits

Appreciation deserves its own column. TI reads it from a mark out of ten in the same survey, shows it by programme, and raises a warning when the show people chose themselves falls below six out of ten.

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

    BBC 'Quality' measures at highest levels in recent years · BBC Press Office, 8 July 2011The Appreciation Index as a score out of 100 from an independently run panel of 20,000 people. Opened 5 November 2026.

  2. 2

    Line of data: TV crunches the numbers · Stephen Armstrong, The Sunday Times, 9 May 2021, archived copyHow the BBC panel marks programmes out of ten and the staff guidance bands it reports. Opened 5 November 2026.

More from the library