How much does advertising matter anyway?

Our results reveal substantially smaller advertising elasticities compared to the results documented in the extant literature, as well as a sizable percentage of statistically insignificant or negative estimates. If we only select products with statistically significant and positive estimates, the mean and median of the advertising effect distribution increase by a factor of about five.

That is from a new paper by Shapiro, Hitch, and Tuchman, via Paul Novosad.


Don Draper stirs in his grave.

Did we just automate the advertising executive?

This is for TV advertising which is rather inert and not the more controversial tech company surveillance style advertising which unfortunately has been proven to be much more effective.

If you know the literature on advertising effects this is not surprising but it doesn't mean advertising isn't valuable. Aggregate sales effects are not the same as effects on the consumer. Most of the value of advertising is from longer-term effects, the building psychological and cultural moats. For the comprehensive explanation read How Brands Grow by Byron Sharp (for a precis, my FT article, How the Mad Men Lost the Plot).

+1 If you look at how corporations are allocating their advertising (more generally "promotion" dollars) you can see the trend quite clearly. For consumer products, instore slotting allowances, and for online, online advertising, are the two trends that target consumers closer to the purchase decision.

TV advertising is just a subset. Sports TV advertising is another set all by itself.

You might want to look at Geskey, Media Planning and Buying in the 21st Century: Integrating Traditional and Digital Media for the trends by each channel and valuation.

Here's a classic study from 1995:

How T.V. advertising works: A meta-analysis of 389 real world split cable T.V. advertising experiments.

Journal ArticleDatabase: PsycINFO

Lodish, L. M., Abraham, M., Kalmenson, S., Livelsberger, J., Lubetkin, B., Richardson, B., & Stevens, M. E. (1995). How T.V. advertising works: A meta-analysis of 389 real world split cable T.V. advertising experiments. Journal of Marketing Research, 32(2), 125-139.
"Conducted a meta-analysis of 389 BEHAVIORSCAN® matched household, consumer panel, real world split cable TV advertising experiments. The study sponsors (packaged goods advertisers, TV networks, and advertising agencies) filled out questionnaires on 140 of the tests, which test common beliefs about how TV advertising works, to evaluate strategic, media, and copy variables unavailable from the BEHAVIORSCAN® results. Although some of the variables did indeed identify TV advertising that positively affected sales, many of the variables did not differentiate among the sales effects of different advertising treatments. Changing brand, copy, and media strategy in categories with many purchase occasions in which in-store merchandising is low increased the likelihood of TV advertising positively affecting sales. The data also suggest different variable formulations for choice and market response models that include advertising. (PsycINFO Database Record (c) 2018 APA, all rights reserved)"

In other words, it was surprisingly hard to move the needle with TV advertising for existing supermarket brands.

I haven't really paid attention to this question since the 1990s. Have there been new developments in thinking in this century

Is this about individual products or about brands? I think of expensive advertising as a form of signalling - it is not about convincing someone that your product is the best, it is about showing your willingness to maintain standards because of the investment. So Coca-cola might advertise a new flavor of coke, but not see much of an uptake (showing up as a null result in this analysis), but their brand image got reinforced, hence an easily replicable soft drink continues to lead market shares a century after it was introduced.

The advertising sector has not been much of a growth sector, its share of the total economy stable or even falling. Of course, what's happened is that much of advertising has shifted from old media to new media (digital advertising), with Google and Facebook capturing most (between 60% and 70%) of the revenues from the latter. Advertising is manipulation by another name. Until Rene Girard, few people understood why advertising works, and fewer still how advertising could be used and abused for ends besides marketing a product. Sure, many people recognize the name Joseph Goebbels, but associate the name with Nazi "propaganda". But advertising is propaganda. Today, propaganda (advertising) is much more sophisticated, thanks to Girard and the quants who know who we are. We are a media obsessed culture, consumers of a constant barrage of propaganda and manipulation, unaware of its influence. Goebbels would envy his disciples.

But this seems to be implying propaganda is as much self-manipulative as it is manipulative. Advertising thinks it is manipulating people into buying the stuff advertised but it also manipulates itself into believing it works better than it does.

I wonder if there's something akin to Godel's Incompleteness Theorem at play here. Propaganda's effectiveness has a fundamental limitation in that it distorts the sponsor's view of reality as well as the target audience.

For example, when North Korea plays up the 'glorious leader' as some sort of god, for whom the birds all sang praise in Korean on the day he was born, is this propaganda manipulating the people of North Korea anymore or manipulating the leadership of North Korea?

So basically advertising works for things that people really need or want. People are rational after all.

Making an intelligent decision in the purchase of a new automobile requires recognizing how it fares among dinosaurs.

Prisoners' dilemma game perhaps? (All companies struggle for the same 100% of the market share)

Based on surveys of adults, Televison effectiveness still ranks higher than other media.

Based on surveys of adults who were asked, Television ranked above other media on effecitveness:

Learning about Products: TV 53%; Radio 5%; Newspapers 10%; Magazines 16%; Internet 17%

Most persuasive: TV 70%; Radio 6%; Newspapers 11%; Magazines 9%; Internet 5%

Source: Geskey, referred to in earlier comment above.

Radio ads must work, because as soon as NY State announced an extended time period to sue alleged molesters, there were lawyer ads galore.

And from what I hear, radio ads must also work if you need to get out of your timeshare or reduce your delinquent federal taxes.

You seem to have mispelled one of the author's names. It should be Hitsch.

What is reputation risk? Does marketing form expectations? What is the cost of failing to meet those expectations? Most old products seem to have a need to keep their names out there. If coke did no advertising on old products, would it lose market share? Advertising agencies convince them they will. Does Berkshire Hathaway advertise?

Wtf does that sentence even mean? Can someone answer in understandable language "does advertising matter" or not?

Never mind. Read the articles conclusion. There, it's much clearer wtf they are trying to say.

Let's call this the Don Drapper Paradox:

Either advertising works or it doesn't.
If it works, equilibrium would be for those selling advertising to use it to manipulate you into buying more advertising. Hence it won't work, at least for those seeking to control society through advertising.

If it doesn't work, then it doesn't work.

Comments for this post are closed