We recently launched Ebiquity and Gain Theory’s new study of advertising effectiveness, “Profit Ability: the business case for advertising”. You can read all about it here.
The study, for the first time, provides a comprehensive breakdown of how advertising across different categories delivers profit across the short and the long term, the likelihood of achieving profit and how this varies across media channel.
It also details the volume of profit delivered by channel, and that’s what makes it different – there is no focus on pure ROI numbers in a vacuum, useful though they are as a measure of efficiency. This is about the bigger picture. It proves what advertising does to the bottom line.
Its findings will hopefully be a marketing foot in the door of the finance department and the C-suite generally. It is also another shoulder to the tiller trying to turn advertising away from its value-destroying course towards short-termism, as starkly highlighted by the recent Binet & Field study for the IPA, “Media in Focus”, and Enders Analysis’s study for Magnetic, “Mounting risks to marketing effectiveness”.
“Profit Ability” is a very complex study and took a year to complete. You can watch videos explaining the methodology here. For such a complex study, it isn’t surprising that we have been asked for more information about some of the findings and how Ebiquity / Gain Theory arrived at them. Now the dust has settled after launch it makes sense to gather the questions together in one place and answer them. Here goes…
Can you give us a handy chart with all the main findings in one place?
OK, no one has really asked for this, but it is a good way to set the scene:
Now, on to some real questions…
How representative of the range of different advertisers is the study?
The findings are representative of the average performance of larger, better known advertiser brands whose media is bought by the major media agencies and is professionally audited by Ebiquity and/or Gain Theory.
As it was Ebiquity’s client base that was used to determine the short-term ROIs and ad-generated profit volumes, which Gain Theory then built on to determine long-term impact, it makes sense to detail the annual revenue distribution of Ebiquity’s client base. It is as follows:
So, all sizes of advertisers are accounted for, with a skew towards bigger companies. This is not surprising if we are realistic and pragmatic. To produce a study like this we are reliant on aggregating the data from advertisers who commission econometric analyses of their campaigns. As a result, any selection bias comes from the fact that the research sample necessarily only includes brands who are dedicated to effectiveness enough that they invest in this type of analysis. As Tom Roach, Managing Partner of Effectiveness at BBH London, has commented: “If they’re paying good money for econometrics they’re probably doing lots of other things right”. The selection bias may be that these companies are good at marketing.
Is it relevant to smaller advertisers?
See above for the revenue distribution of the companies involved. However, we can assume that most smaller advertisers one day want to be bigger. So, this research offers directional guidance on what big advertisers do, and what works.
Are all types of advertising represented?
Demand-generating forms of advertising are represented, defined as:
- TV advertising: spot advertising (1,280 campaigns)
- Print advertising: print newspapers and magazines (980 campaigns)
- Out of Home advertising: 6 sheets, 48 sheets, 96 sheets – standard & digital formats (580 campaigns)
- Radio advertising: spot advertising (540 campaigns)
- Online Display advertising: online static display, e.g. banner ads, incl. Facebook static ads (330 campaigns)
- Online Video advertising: Broadcaster VOD, social VOD – e.g. Facebook - & YouTube (158 campaigns)
Why wasn’t online search included?
Ebiquity/Gain Theory did not include online search as it is a demand-harvesting rather than demand-generating investment. For the same reason, the value of high street location or prime shelf space in supermarkets is not included in this study. That said, Gain Theory did note that they could find no examples in their database of online search having a long-term multiplier effect.
Why isn’t social media included?
It is, however there isn’t the data available yet to treat it as a separate medium. Ebiquity calculated that the average short-term profit ROI for social media was £1.14, so in between Online Display and Online Video. This make sense as there will be social media advertising in both the Online Display and Online Video datasets. However, as this calculation is based on a lower than ideal number of data points, this figure for social media ROI should be treated directionally rather than conclusively.
Why does online display perform so badly?
There is no way to sugar this pill; the data shows consistently that online display performs poorly. When Ebiquity looked at the range of short-term ROIs generated by online display, they found that 37% of online display campaigns delivered some profit and that the highest ROI recorded was £2.38 (compared with a high of £6.12 for radio, £5.07 for print and £12.96 for TV). Gain Theory found no long-term multiplier effect for online display. So, it is relatively weak in the short term and does little in the long-term.
As a media channel that is predominantly used for response advertising it’s surprising that it doesn’t pay back in the short term. However, the likely reasons behind this have been well reported: issues around ad fraud, viewability and arbitrage are rife and this article on the minimal impact of P&G’s move away from online display supports Ebiquity’s findings.
Why doesn’t the study split out different forms of Online Video into BVOD, YouTube, Facebook etc.?
There is not enough data on each individual form of Online Video from which to draw solid conclusions. The data used within this study is based on client-commissioned analysis of their own campaigns’ performance and very few clients have online video channels separated in their analyses (they tend to be planned together and account for relatively small volumes of spend making it harder and less cost effective to separate out through econometric analysis).
However, previous Gain Theory research into the performance of online video commissioned by Videology demonstrated that campaigns with a higher proportion of Broadcaster VOD in the mix delivered a stronger ROI – i.e. the higher the quality of your online video (high view through rates / full screen / sound on / quality brand safe environment) the better the ROI.
These findings likely explain the broad range of ROIs delivered by Online Video in the Ebiquity/Gain Theory study; campaigns with a high proportion of quality online video perform well, campaigns with a low proportion do not.
Did the study include cases where an advertising effect couldn't be measured?
‘No reads’ were excluded from the analysis unless it was concluded that no read meant no effect. Ebiquity often combine holistic econometrics with other methodologies to provide a total view of performance – a good example of this is geo-testing where they exploit local weight variation. This can be useful for hard-to-measure channels/campaigns.
Is this a planning tool?
No, it is a mathematical basis upon which to make better-informed media planning decisions. It is a starting point from which to assess how much weight or consideration to give different advertising channels before moving on to the detailed planning required to meet specific objectives.
TV does very well – but then it would wouldn’t it, given who commissioned the report?
In the words of Ebiquity’s Andrew Challier, the findings are “objective, wholly independent from Thinkbox and not designed to prove a particular point of view. There was no original research in the traditional sense; it came from us slicing and dicing our extensive body of work. By asking us and Gain Theory to compile this report, Thinkbox had to accept there was a chance the result may have not been what they would want to hear.”
Obviously, we are happy with what we have heard – but not just for TV advertising, for all the forms of advertising that have had their profit ability proven in this study.
If you have any more questions, please do get in touch.