Coca Cola could have saved themselves millions in airtime costs for Christmas advertising


[Dashboard view of all the filtered demographic responses; the colour of the filter box for each demographic marries with the colour of the lines on the ‘Viewer Graph’]

What is Sensum?

Sensum’s audience insight system provides ad agencies, market research firms, and broadcast media with an innovative way to accurately measure customer response to media.

[The 3 parts to Sensum : wireless finger sensor, mobile device with Sensum app installed, and dashboard for viewing, filtering & reporting results]

Current market research relies on focus groups’ verbal or written responses to questions before and after viewing media. This qualitative approach relies on cognitive processes where responses are consciously formulated after the event. These responses are prone to self-editing and test biases that skew or distort the data, including moderator bias where the presence of a moderator or the test environment itself influences responses

Recent research in neuroscience has suggested that up to 95 percent of our purchase decision making actually takes place in the subconscious mind. Strong brands, e.g., excite parts of the brain associated with pleasure and reward and trigger physiological reactions.

As physiological responses are involuntary, instantaneous, and may be measured quantitatively, they provide a better measure of the level of engagement and emotional response to stimuli while avoiding conscious cognition test biases.

With Sensum both non-conscious and conscious cognitive responses can be recorded, mapped, and compared, providing deep insight into how audiences truly respond to the media. This allows development of media that is both more engaging and cost effective, resulting in higher return on investment for brands, advertisers and content producers.

What did we test?

We selected a commercial that would push emotional buttons – one of the Coca Cola Christmas commercials.This was not a test in association with Coca Cola but a test that we at Sensum ran to demonstrate the platform’s capability.

[Screengrab of the Coca Cola commercial that we tested]

The Coca Cola commercial has successfully become associated with Christmas, and we found that people would say ‘It’s Christmas now that the Coca Cola commercial is on TV’.

We wanted to see if the commercial was effective via an audience’s conscious & non-conscious responses, and to see if any savings could have been made by understanding the effectiveness of the commercial.

The Test

All test subjects were aware that they were going to be watching a Coca Cola Christmas commercial but not aware of which one.We selected one from some years ago so as to avoid people having seen it on TV recently.

We tested 51 subjects from Belfast, N.Ireland, with a spread of ages from 18-81.The exact breakdown of age ranges tested was :

18-24 : 10 people

25-34 : 11 people

35-44 : 8 people

45-54 : 10 people

55-64 : 5 people

75+ : 7 people

We used Android smart phones, with the Sensum app pre-installed, to carry out the mobile testing, and reviewed the results on our platform dashboard.

[The Sensum app and wireless sensor for capturing conscious and non-conscious audience reponses]

Each participant answered questions relating to this test and their responses were uploaded to our secure cloud server. This was the conscious response to the commercial.

And we wirelessly captured the galvanic skin response [GSR] of the 51 participants whilst they viewed the commercial.The higher the level of engagement the higher the GSR measurement and therefore the more effective the media.The media was streamed to the smartphone and the GSR data was then uploaded to our cloud server for review.This was the non-conscious response to the commercial.

The Conscious Response

We asked a pre-question & a post-question to establish the valence, the spread of negative to positive response, from before to after viewing the content.The aim was to see whether the commercial had achieved it’s aim of instilling the Christmas spirit in the audience by making them feel that Christmas had begun.


The question asked prior to viewing the commercial was :

  • Which of the following describes how you presently feel?

Across the 51 subjects 24% felt ‘less than good’, with 59% feeling ‘good’ & 17% feeling ‘very good’ or ‘excellent’.

[Dashboard view of response to ‘pre-question’ asked before viewing the commercial]


The question asked after viewing the commercial was :

  • Which of the following describes what you thought about the media?

Across the 51 subjects only 16% felt ‘less than good’, with 57% feeling ‘good’ & 27% feeling ‘very good’ or ‘excellent’.

[Dashboard view of response to ‘post-question’ asked after viewing the commercial]

So based on the answers of the subjects you can see that the commercial was successful in generating a feeling of Christmas joy, with ‘Holidays Are Coming’, and Christmas themes through the audio and visuals.

You can see that there was a clear shift in percentage ‘feel-good’ when viewing the commercial as the numbers dropped in the ‘less than good’ responses with feeling ‘very good’ or ‘excellent’ increasing.

The Non-Conscious Response

The graphs display the engagement levels of the various demographics that we tested.

[Dashboard view of video & graphing tool for reporting on session results displaying average response & filtered demographic responses]

The orange graph is the overall average response which can be seen in 2 locations on-screen.

The top window is the ‘Viewer’ window where all graph overlays can be applied to the video player; the bottom orange graph displays the overall average response to the full media clip from start to finish so you can ‘quick-glance’ the points of interest.

What is of immediate interest is that the average graph starts high with a slight build then tails off continuously to the end of the commercial. But overall from 28 seconds onwards the engagement levels drop off.

We applied graphing filters to view the responses of the various age ranges of our test subjects, which are displayed with various colours for each range.

The 35-44 year olds [the light blue line] were the most engaged from start to finish with the 18-24 year olds [the lime green line] starting highest and dropping off in engagement quickest.

[18-24 yr olds & 35-44 yr olds highlighted on graph]

From 28-30 seconds you can see the shot of the Coca Cola trucks coming through frame which results in an uplift of most results [this is the still of the video on the graph], and again between 41-43 seconds a long shot of the trucks crossing a bridge that lights up creates a surge in response from most age demographics.

[Coca Cola truck on the bridge at 41 seconds which caused an uplift in audience responses]

The Conclusions

While the traditional survey showed uniformly positive audience response, the physiological response showed that the audience subconsciously disengaged after 28 seconds, suggesting the ad could be cut in half and still achieve the desired response. It is estimated this reduction could result in £10m in airtime savings in the UK alone.

This calculation was made based on the following assumptions :

A commercial will cost between £60-250k to broadcast depending on time of day, so we took an average amount at £150k per 1 minute broadcast slot.We estimated that the Coca Cola commercial was probably screened 4 times per day on 3 channels in the UK for 12 days leading up to Christmas. That is approximately £20m in airtime costs for a 1 minute commercial.

If the commercial had been halved in duration it would have saved £10m in airtime costs.

Production savings could also have been possible by testing well designed storyboards, with different audio applied alongside different durations, with a range of target demographics.

It’s important to say that a 30 second commercial may not have been the most effective either, but it’s clear that the 60 second one wasn’t the most engaging commercial or the best spend of airtime money for a Northern Irish market, and perhaps for a wider UK market.

Testing on other demographics & in other territories may have returned different results.But let’s just say the same results were returned when the commercial was tested in all the countries that Coca Cola is sold.

Imagine what sort of overspend Coca Cola incurred in advertising airtime costs around the world!