The Empirical Generalizations in Advertising Conference brought together an outstanding group of researchers and practitioners in diverse areas of advertising who presented on 44 different papers that summarize rsearch-based knowledge in the field. What works? What doesn’t?

Use the category bar on the right to go directlty to reports on some sessions of the conference (EG Conference Reports) or to more extensive notes on sessions (EG Conference Notes)

What’s happening with brands? Customers are price sensitive; more products are failing; more trips to discount outlets

Changes in promotion spending – it’s up -  advertising has shrunk at the expense of trade promotions.

A longer-term view: looks at “quantity premium” – how much more a brand sells compared to similar brands, and  “margin premium” – Brand marketers want to see a high quantity premium and a high margin premium

Advertising and distribution increases quantity premium; discounting hurts (finding from study that looked at poor performers and good performers).

Strategy and price premiums: distribution and discounting reduce price premium; advertising increases it.

Sum – many brands are flailing, short-term orientation favors discounting, long-term metrics favor advertising

Bruce Tedesco, New modeling methods for simulating advertising impact

Bruce opened with a couple of observations:

1. Advertising analysis requires modeling methods based on dynamic principles

2. Consumer media consumption across channels leads to virtual days exceeding 24 hours

Dynamic models – agent-based models, which have been around for some time but only recently applied to marketing, simulate what happens among consumers in a virtual marketplace. Running not a single simulation, but many, we are able to gain empirical direction from studying the effects and interactions that arise when inputs, like advertising, are varied. Simulations are really about “emergent behavior.” If you’ve ever played the SIMS or a similar game, you’ve had experience with dynamic modeling.

“Rather than calculate,” Bruce said, “we compute. We don’t solve a single equation.” The emergent behavior arising from the simulations informs our analysis and strategy.

Discussion

How do you validate the simulations? There are ways to calibrate the simulations in real time, e.g. using multiple samples of the data. Best have been when we can run real time in market, and we can see that the results of the simulated marketplace match with the actual sales results in the market.

Theories have interesting validations – a lot of work in agent based modeling has been done with modeling the past. All of the conditions in Florence in 1400 were fed into a model, the same type of guilds appeared.

Real validations come in working with clients and sticking with it long enough.

Gian Fulgoni

Display ad click rates under 1%, on search they run about 5%. Sales impact of a search ad is greater than that of a display ad.

Gian reviewed the topic of cookie deletion.

Cookie deletion is a real issue and “renders invalid” most studies that rely on cookie-based metrics – important implications for understanding consumer behavior and for estimating campaign reach and frequency. 30% of Internet users delete their cookies in a month. Deleters do so an average of 4x/month.

Problems: up to 2.5 times overstatement of unique visitors in server logs; up to 2.5 times overstatement of reach and a similar understatement of frequency in ad server logs; one loses ‘visibility’ of machines with deleted cookies, thereby understating advertising ROI.

Gian summarized a number of comScore studies exploring the effects of search and display ads on site visitation and sales.

Display ads impact site visitation. Impact is greatest during the first week, an effect which lasts over the four weeks following the first exposure. This is true across categories.

Display ads build sales – online and offline.

Combination of search and display leads to greatest sales lift. Search and display synergy: 119% sales increase vs 82% for search and 16% for display only.

EG 1: 20% of WOM is stimulated by Advertising. Influencers are 3x as likely to talk about ads.

WOM varies by category: highest is entertainment/movies, technology, personal care & beauty. Least: Health/healthcare

EG 2. Ad-influenced WOM is about 20% more likely to include active recommendations to try/buy product.

Implications:

The rise of wom does not portend “the end of advertising.”

The target for advertising needs to change (influencer-based media plans)

Messaging might need to change (from persuasion, targeted to prospects, to messages that foster brand advocacy, targeted to current customers).

Question: do you break out your results by social network users? We find that people who are conversation catalysts participate online and offline. They’re people who like the social interaction.

Is there a difference between offline and online wom? Online wom – about 1/2 of it is teenagers. From a marketing standpoint, need to keep that in mind.

Distribution of wom conversations by channel? 75% is face to face; 15% is phone; 3% texting; 3% email; 1% chat rooms. About 3.5 billion brand conversations per day.

Mark Earls

We need to develop new approaches to understand consumers, why? Three reasons:

. Perhaps advertising works through people.

. Internet is not media, it’s communications based not a message broadcaster.

. Only in the Anglo-Saxon world do we conceive the audience as being comprised of individuals.

Unless people can see what other people are doing, behavior doesn’t spread. People copy patterns.

Random copying – long tail, NetFlix. Random copying is stochastic. Selected items are copied differently.

Advertising works on top of and amplifies cultural tides.

Message Characteristics: Shoppers are most responsive to “news” and least responsive to traditional brand messages.

Product: category is a key factor – food/entertainment have the highest lift.

Moderating Variables: need state of shopper (message receptiveness varies by time of day and day of the week; quality and frequency of exposure). Where the sign is located, its orientation and visibility affect shopper response.

Digital signs, whose messages could change by time of day, lifted store sales.

In an experiment with Eddie Bauer, store traffic increased 23 percent and sales increased over 10 percent when paper signage was replaced with digital signs that change during the day.

David Burgos, Old thinking —> new thinking = multicultural mindset

Multicultural concern is critical because US population is multicultural and influencing the wider society – foods we like, politics, and so on.

3 Principles:

1. Basic advertising principles are generally valid across cultures

2. Culturally targeting you advertising tends to work better than non-targeted communications

3. didn’t have time for the 3rd one.

Comments from the floor:

The concept of multicultural is problematic because it masks the differences within groups.

Let’s accept multicultural when relevant, and let’s distinguish those areas where multicultural is the same as the wider society. For example, studies of Hispanics show that “family” is very important, but it’s also important to many other groups.

J. Walker Smith and Bill Moult

Their study asked people about ads they saw or heard recently that had an impact on them.

3 Generalizations:

1. Recently recalled ads were more likely to have left a positive impression than if they were on digital media. “Ads in traditional media kick a$$.”

2. Prior awareness of the brand is related to positive impression of the ads. Key implication – you can improve the effectiveness of ads by doing things to help make people aware of the brands.

3. If the recent ads recalled left a positive or negative impression, respondents were likely to exhbit certain negative or positive behaviors, but not if the impression was neutral.

Began by describing market changes that may influence the patterns of elasticities observed from an earlier highly regarded meta-analysis by Assmus. Research found that they supported five findings from Assmus, and 7 new ones.

Other things being equal , advertising favors durables and growth products.

Confirms a finding talked about across the two days, that the average elasticity is .1., and that the long-term effect is about twice the short-term impact.