Data Spotlight: How Online Brand Advertising Can Influence Every Step Along the Consumer Purchase Funnel

The Internet can have an impact at every phase along the consumer purchase cycle, from creating initial awareness and interest for a product, service or brand, to stimulating purchases, to delivering post-sales support and reinforcing brand loyalty. But what about the measurement of this impact?

As a crude indicator of how marketers generally value online media, 36% of advertisers in a survey by Morgan Stanley said the Internet was “effective”—considerably higher than for any other media. It was not clear from the survey, though, whether marketers were thinking primarily of search, display ads, their own company Websites or other online tactics.

As validation, a more recent ANA study released in May 2009 found that while television was rated the most important media channel in terms of effectiveness (64% of respondents), the Internet was a close second (61%).

How Display Ads Impact Brand Metrics

Dynamic Logic, InsightExpress and other survey firms have conducted countless studies—covering all types of product categories and services—to measure the brand impact of online display and video advertising.

The latest aggregated data from Dynamic Logic, based on 2,380 campaigns measured through Q4 2008, found that those respondents exposed to online advertising registered slightly higher increases in key brand metrics, on average, than those who were not exposed (that is, the control group). The delta lift (expressed as a point difference) for exposure to online ads was 2.4 for aided brand awareness and 2.6 for message association.

For additional information on the above chart,
see Endnote 103522.

But the above figures are based on across-the-board averages. Importantly, the creativity and contextual relevance of ad campaigns can heavily influence the results. When Dynamic Logic dissected the data, top-performing campaigns boosted key brand metrics by significantly higher levels than did the lowest-performing campaigns.

For example, aided brand awareness increased 8.9% for top-performing campaigns, versus only a 2.4% increase on average and a drop of 2.3% for the lowest-performing campaigns. Similarly, top campaigns pushed up purchase intent by 7.1%, versus only 1.3% on average.

For additional information on the above chart,
see Endnote 103523.

Dynamic Logic also compared campaigns for different categories of products and, again, found highly differentiated results. For example, on the measure of boosting purchase intent, entertainment, consumer packaged goods, home improvement and telecommunications categories performed much better than average, while apparel, restaurant and retail did relatively poorly.

When it comes to measuring the efficacy of display ads on branding metrics, context appears to matter. According to a study by the Online Publishers Association (OPA) and Dynamic Logic, display ads within branded content sites (for example, NewYorkTimes.com, Weather.com) do a better job of boosting brand metrics than when those same ads are placed within more generic portals or ad networks.

For additional information on the above chart,
see Endnote 102827.

Multiple studies prove, too, that online video ads can have an outsize effect on standard brand metrics such as awareness, message association and purchase intent.

Integrating Search and Display Ad Measurements

Intuition (and plenty of data) tells us that search works best when it is complemented by online branding efforts that create awareness, interest and desire among prospects.

First, as consumers ourselves, we realize that often the very idea of searching for a particular product or brand comes from having seen advertising. Second, when we are looking to possibly purchase something and conduct a search inquiry, there is a natural tendency to click on those search links that represent brand names we know and trust, and to disregard unknown brands.

Most agree that online display ads can act as a stimulus for driving searches for brands or products—which of course is a key step leading to purchase.

“There’s an image, a perception, a value that [consumers] have at the point that they conduct a search. You know, it’s not like consumers are sitting there with no impression of a brand, until they conduct a search,” said Gian Fulgoni of comScore.

“If you see a display ad, there’s an incremental impact, but it’s not as great as the incremental impact of a search ad,” he added. “If you combine the two, you get synergy and the combination is greater than the sum.”

In an older yet still relevant study by the Microsoft-owned Atlas Institute, 11 online advertisers were evaluated by measuring both their display ads and sponsored search clicks. In the study, Internet users exposed to both search and display ads converted at a much higher rate than did those exposed to search or display alone.

Similarly, a Specific Media survey found that online display ads can boost search activity for many product categories, especially for travel and tourism, health, personal finance, automotive, news and media, and property and real estate; the average lift was 155%.

In a January 27, 2009, interview with eMarketer, Beverly Thorne, senior vice president of marketing at Century 21 Real Estate LLC, explained the search/display connection for her brand as follows: “Our own empirical results showed us that our online investments were performing substantively better at generating leads. From December 2007 to December 2008, we improved the efficiency of our lead generation by reducing our cost per lead over 60%. At the same time, we multiplied our number of leads by over 235%.”

The mastery of Web analytics plays a key role in tying display ads to search results. According to analysis by Steve Kerho, VP of analytics at Organic, display ads can increase a search ad’s click-through rate by 25% to 30%. Until recently, though, this influence of display ads has typically not been measured by advertisers.

How Display Ads Drive Site Traffic

Advertisers often have the goal of driving qualified traffic to their corporate or brand Websites. Based on numerous independent studies, display ads can help get this job done.

For example, comScore evaluated 139 studies linking display ads to site visitation levels and found that the average lift in the number of visitors to the advertiser’s site—comparing visitation levels of exposed and nonexposed groups—was 65% during the first week following the first exposure to an ad. Additional, though somewhat muted, lifts were seen in the following three weeks.

In another industry-specific example, a joint study by comScore and analytics firm Crossix Solutions, in partnership with Yahoo!, evaluated the ability of display ads to drive consumers to pharmaceutical advertisers’ Websites. After seeing online display ads, consumers were more than three times as likely to visit pharma advertisers’ Websites and twice as likely to fill a new prescription. In addition, exposed consumers were 92% more likely to search for trademark names and phrases compared with those who did not see the display ads.

The study was conducted among a base of 73 million consumers and concluded that the display ad campaigns provided an estimated return on media investment of 3X. Not a bad ROI.

How Display Ads Impact Online Sales

For most categories of products and services, consumers rely heavily on the Internet to do research, or “window shop,” prior to making the final purchase. Given this proclivity, marketers have an obvious opportunity to influence the awareness, preferences or even behaviors of these online shoppers—at the precise time when they are in exploration or consideration mode. In the 90 days leading up to a sale, consumers see an average of 18 ads for a product, according to Microsoft’s Atlas Institute.

eMarketer estimates that 152 million people, or 86% of Internet users ages 14 and older, shop or browse online for possible purchases that may or may not occur online. Of course, the tendency to shop on the Web varies by category. A worldwide survey from Universal McCann indicates that holidays/destinations, consumer electronics and travel are the most-shopped categories online.

A study by Penn, Schoen & Berland Associates found that US online shoppers would research high-consideration products and services, such as cars, computers, new doctors, vacations and mobile phone plans, prior to making an actual purchase.

But less expensive products are increasingly researched online as well. According to Jeffrey Grau, eMarketer senior analyst, consumers today access the Internet to look up even relatively low-interest products such as shampoo. Said Mr. Grau, “Consumers might go to a shampoo manufacturer’s site to learn whether a product contains paraben preservatives, or is tested on animals or made from animal ingredients. They also might visit a retailer’s site to read customer reviews to hear from others whether a shampoo is really unscented or causes scalp irritation. Consumers also go to coupon sites to get additional savings on a shampoo purchase.”

Display/branding ads can also influence consumers to the point of purchasing products online, though rarely is the impact in the form of an immediate click.

In several studies, comScore evaluated the combined influences of search and display ads on consumer online buying behavior. Search, given its obvious indication of purchase intent (that is, those who take the time to search for a product are usually in-market), has a stronger influence on consumer buying behavior than display ads alone. But when both search and display ads are combined, the overall impact is significantly greater than that of either search or display ads individually.

For example, while display ads alone provided a 42% lift (test versus control) on the percentage of consumers making a retail purchase online, and search generated a 121% lift, the combination of search and display ads together produced a 173% lift. Further, the one-two punch of search and display resulted in significantly higher dollar spending per thousand consumers exposed.

How Display Ads Influence Offline Purchases

“The current lack of visibility into offline purchasing [as a result of online display advertising] consistently leads to dramatic underestimation of display advertising ROI.”
—comScore white paper, “How Online Advertising Works: Whither the Click?” December 2008

Unfortunately, while online advertising can have a substantial impact on offline sales, most marketers fail to measure this important connection.

According to McKinsey’s June 2008 digital advertising survey of 340 senior marketing executives worldwide, only 30% said they even considered the offline impact of online marketing. However, marketers that did look at those metrics were more satisfied with their online efforts and said they planned to increase spending on them by 38%.

To be fair, integrated measurement systems are often difficult and pricey—but in the long term, for marketers that measure search and display advertising together, the results can be particularly encouraging. A September 2008 comScore study found that by using search and display ads in combination, marketers can significantly boost the dollar value of offline retail sales versus using either search or display only.

As evidenced by a separate survey conducted by comScore Networks and Yahoo!, the search/display combination of ad messages increased in-store purchases significantly. In the study, those online shoppers who had seen both ad types were 43% more likely to be converted to in-store buyers. In comparison, consumers who had seen only search ads were just 26% more likely to be converted, while exposure to only display ads lifted conversion by just 6%.

As a rule, the impact of online research/shopping behavior is greater on in-store sales than Web-based sales. According to comScore, for retailers with online and offline sales channels, approximately 68% of the impact of display ads was found in the offline channel.

In the aggregate, Forrester estimates that in 2009, Web-influenced or cross-channel store sales will reach $758.8 million, dwarfing the sales transacted directly online. In other words, the Web’s influence on sales is much greater than its direct e-commerce impact.

Social Influence

Although there is only limited evidence at this point, a few studies suggest that advertising in social media can affect offline sales, too.

A partnership research study involving comScore, dunnhumby and MySpace was conducted to track offline purchases of Internet users exposed to ads for a personal-care brand on MySpace. Of the nearly 80 million people exposed to the campaign, fewer than 1% visited an advertiser page on MySpace, though about one-half who did also visited the personal-care advertiser’s site.

Despite this small percentage of visitors, the campaign met its ROI objective. According to Advertising Age (April 13, 2009), “[The campaign] produced $1.28 million in offline sales, as measured by dunnhumby, which compared purchases among shoppers not exposed to the campaign with purchases among those who were.” This resulted in an impressive 28% return on marketing investment.

In sum, marketers who fail to assess the cross-channel impact of their online branding efforts—particularly in this economic climate—are significantly underestimating their effectiveness. Leading marketers are developing tracking mechanisms that allow for comparisons between offline and digital marketing. And they do not have to be complicated or expensive. For example, in order to track sales across channels, retailers can use online coupons that are redeemable on Websites and in stores.

“When you start to link in-store purchase behavior with online advertising, it’s incredibly valuable,” said Mr. Lanctot of Razorfish. “[It] can help marketers understand how online marketing drives offline behavior.”

Three Factors for Online Branding Success

You can have all the precise numbers, measurement tools and metrics at your disposal—and completely bomb with your online campaign if your creative messaging is substandard or badly targeted.

1. Don’t Ignore the Creative

Under the pressure of marketing accountability, and in our zeal to quantitatively measure everything that moves, marketers must not abandon creativity as an essential variable in online branding success. Measurement tends to look backward like a rearview mirror, while creativity seeks to shine the headlights on a future attitude or action—on the part of the consumer marketers are trying to reach and influence.

As Jon Gibs from Nielsen Online said, “Creative is about 70% to 80% of the effectiveness of advertising.”

“If your creative isn’t good, then you aren’t going to create engagement.”
—Jeff Marshall, managing director, Pixel, a digital creative agency owned by Publicis Groupe, as cited in The Wall Street Journal, May 6, 2009

Ken Mallon of Dynamic Logic, whose company has evaluated the brand impact of thousands of online campaigns, was very vocal about the need for good creative: “By far the biggest driver of brand impact success is the creative. The best ads that we see in terms of performance online tend to be ones that almost have a magazine feel. They look nice; people think about things like having the right human form in there, the right product shot.”

Effective creative can be planned, according to Mr. Mallon. “We have a set of 10 standard best practices. The clients that follow those do really well. But most people violate one or more of those routinely, and that’s why online advertising just isn’t that great on average.”

Some in the industry go so far as to say that the creative form of online advertising is not only critical, it transcends the measurement issue entirely. In eMarketer’s interview with online publishing pioneer Martin Nisenholtz, it was clear that finding ways to make online ads more emotionally evocative is paramount to branding success.

“I don’t think the measurement issue is holding back brand dollars online. The bigger concern is that the Web still hasn’t found a way to create the kind of emotional involvement that television creates for people,” said Mr. Nisenholtz. “I can pretty much promise you that television didn’t evolve as a dominant brand-building medium because somebody started with a measurement. It evolved because marketers said to themselves, ‘This is an incredibly powerful way for us to communicate and transmit the emotive powers of branding.’ And then they figured out how to measure it. In other words, measures are the tail, not the dog, and the dog hasn’t yet been invented online. So we have to invent the dog.”

Also speaking from the online publisher’s side is Pam Horan, president of the OPA, who feels there has not been enough focus on the creative process and making online ads that engage with consumers. “Creative is a big issue,” she said in an interview with eMarketer. “At the OPA, we’re trying to spur a creative renaissance, so to speak, to show the opportunity for online advertising to deliver a rich experience, tell a unique story and a brand-oriented message.”

2. Size Matters

When it comes to banners and other forms of display ads, larger, more intrusive ads tend to perform better than smaller ones, on average. Dynamic Logic and InsightExpress studies have proved this correlation.

The size and shape of display banner ads can also affect engagement results, at least as measured by time spent. According to a time-exposure study conducted by Lotame Solutions, the large 300×250 rectangle caused Internet users to spend significantly more time with this unit versus more narrow (and creatively limiting) ads, specifically 728×90 and 160×600 banner ads. Consumers spent over 13 seconds with the rectangle (300×250), more than double the next-best-performing ad unit, the 728×90. The large size and shape of these rectangles, which end up smack in the middle of valued content, means they are difficult to ignore.

Taking such data into account, Web media publisher MSNBC.com, for one, is launching a new Website design that features larger-size ads, as well as exclusive sponsorship placements, to attract budget-conscious advertisers.

Furthermore, as one in a series of solutions to the creative challenge, the OPA recently released three new innovative online ad units designed to create brand experiences—within the ads. “The idea is to deliver the brand experience right on the pages of these rich content sites [e.g., OPA publisher sites] so people won’t have to click away,” said Ms. Horan. “These units are much larger…. The marketer’s share of voice increases because the consumer is being exposed to a bigger ad for a longer period of time.”

3. Focus on Targeting and Relevance

The Internet, partly because it is so fragmented, allows marketers much better opportunities to finely target their messages to the right individuals, and often at the right time and place. The “spray and pray” approach may have worked in the past for traditional media campaigns, but it most certainly does not work on the Web, where consumers almost expect ad messages to be more relevant. We have the targeting technology, but we need to use it.

“Don’t count the people you reach, reach the people that count.”
—David Ogilvy, world-renowned advertising executive

Studies from Dynamic Logic and InsightExpress, for example, show that contextual placements, aligning brand messages that are in sync with the surrounding editorial content, tend to be more effective for highly targeted categories, such as autos, pet care, baby and pharmaceutical products. Many other studies support the ability of contextual placements to boost results. In that regard, online is no different than other media.

Another targeting technology experiencing renewed interest among advertisers is behavioral targeting. With behavioral targeting, ads are served up to Internet users based on the past surfing behavior of the individual consumer. Conceptually, this means that advertisers are buying audience, not Webpages or impressions.

This is good for the advertiser, since it makes for a more efficient media buy. It’s also good for the publishers, since they can sell advertising space that might have gone unsold. And finally, it is potentially good even for consumers, since they are more likely to see ads that are relevant to their interests.

Because behavioral targeting has the potential to efficiently get the “right” ads in front of presumed interested parties, much like search ads, many in the industry see a resurgence coming in this type of technique.

“These improvements [behaviorally-targeted banner ads], driven primarily by better targeting, will also likely boost aggregate spending, as advertisers moving online begin to get the same tracking and metrics that have made search advertising so appealing.”
—Ned May, director/lead analyst, Outsell, in BtoB magazine, April 6, 2009

Forrester Research currently estimates that nearly one-quarter (24%) of online campaigns rely on some form of behavioral targeting data. This was substantiated by a February–March 2009 survey by Forbes.com, which reported that 31% of senior-level marketers were using behavioral targeting. Finally, Datran Media surveyed marketers and found that 65% either already used or planned to use behavioral targeting in the future.

Of course, behavioral targeting, which relies on cookie technology, has two serious obstacles:

Between 30% and 50% of Internet users regularly delete their cookies, rendering behavioral targeting virtually useless for these consumers.

Mounting privacy concerns could end up derailing the use of behavioral targeting, or at least seriously limiting its impact.