This article is only available to paid subscribers.
Over the past nine years, paid search (PPC) and search engine optimization (SEO) have taken their rightful place with seats at the table where major marketing budgeting and planning decisions are made. PPC, with its solid ability to convert visitors and deliver consistent ROI even with increased competition, may have got to the table first, but SEO is now regularly proving to enterprises that the more you can invest, the greater the return.
We know that PPC and SEO work well together, but what about when you throw in other marketing initiatives and measure the reciprocal effects of search to online display advertising (ODA) or offline campaigns? The marketing industry still struggles here, but many signs point to increased understanding of the dynamics. ODA and online public relations (PR) are already being measured together with search, and many ideas and tests are underway to better tie offline with online.
Ironically, technology is a little behind the needs and wish lists of marketers hooked on very advanced analytics systems. Companies are forcing the entrenched leaders in web analytics to focus on increasing research and development. Those with access to this type of data naturally want to be able to dig even deeper than before.
Display And Search – Who Gets The Credit?
In 2006, when Avenue A | Razorfish released their “Actionable Analytics” report, they made official what many in the online marketing industry already knew — display advertising does affect search. The bigger statement? Display also helps to increase conversions. The problem was that there was no industry-accepted way to divide the credit between display and the final click, let alone any other steps taken along the conversion path.
Yahoo! announced on their blog in November 2008 that they would better track the relationship between display and search, and formally assign the “assist” metric to deserving events such as impressions and view-throughs. They said, “Assists measure the total number of times that display ads or search keywords contribute to the conversion of another ad or keyword. Combined with conversion data, the assist reporting provides a full picture as to the performance of online campaigns.”
The problem of attribution modeling remained solely the responsibility of individual marketers, however, leading to fears of discrimination or bias against some marketing functions. For example, what if a CMO was partial to display, having dedicated 75% of his or her budget to that function, and wanted to assign a greater attribution weight to a view-through? Would the search team’s ROI or ROAS suffer?
To investigate further, Rosetta (the company I work for) has partnered with ClearSaleing to better understand what web analytics calls the “purchase path.” The system we’ve developed measures multiple touch points with a site or its campaigns through to eventual purchase or other conversion. When our search clients have also run ODA with us, we set forth attribution percentages based on a variety of factors. In many cases, clients have been more willing to assign a high conversion attribution percentage (up to 75%) to search as a first-measured instance than to ODA.
ODA still gets full credit when someone clicks on a display banner and converts without any search behavior having been measured; however, it cannot claim full credit and ROI unless it was the sole touch point. Of course, the same thing occurs to search campaigns that used to get all the last-click credit. When averaged out, ROI or ROAS continues to be strong for ODA, PPC, and SEO when looked at in aggregate across clients.
One of the most important SEM lessons to learn from conversion attribution modeling is that last-click modeling can actually be dangerous to the advance of an overall marketing campaign. In the past, many well-known brands shifted money away from non-branded “head” search terms because so many conversions occurred with a last click that included a brand or product-inclusive search query. Purchase path analysis continues to demonstrate that this is faulty decision making, done without full insight.
Both PPC and SEO consistently drive traffic to websites for broad head terms that end up converting after a branded search. Thus a decision to cut out broad PPC, or to focus fewer resources towards SEO for competitive head terms, could end up diminishing branded conversions. Another bad decision may be to cut out bidding on PPC for branded terms. Although the brand may rank first organically, having both PPC and SEO gives consumers an extra nudge of confidence when they finally decide to act.
ODA should drive branded search terms, and it continues to perform, especially when keyword-targeted ODA is employed. Targeting visitors with insurance banners following insurance-related searches continues to be an effective way to drive overall conversions that can be attributed partially to ODA and partially to PPC. But how does a big TV or print advertising campaign affect this?
Search And Offline Media – Ships Passing In The Night?
Many Internet-focused marketers wait patiently for the money shot whenever they see a major brand they are working with on TV, in a magazine, or in the newspaper. For them, the money shot is a mention of the domain name either by the voice narrator or printed on the screen or in a magazine. Unfortunately, SEMs consistently are let down. “Why didn’t they mention the domain?” is a familiar angst-ridden thought when the 800 number or local address is mentioned.
By not mentioning the URL in other advertising, companies sabotage their online marketing teams. Although most major brands have the same root domain as their brand name, people still search, opening the door to competition. Bing, the new Microsoft search engine, may cause a shift in this practice of foregoing URL mentions, since their interface actually calls out direct competitors even when conducting a brand name search.
Two main reasons support brands trying harder to drive direct traffic to their domains URL inclusion in offline ads. First, people actually still type the brand into Google, fearing that they may end up on the wrong domain should they try to navigate directly. Second, direct traffic doesn’t always convert on the first try, so if the consumer enjoys the user experience and is provided with memorable calls to actions or benefits, he or she may end up using a search engine to find the site again.
Even though the predominant two-step purchase path in SEO analysis is where the first step is broad search and the second step is a direct visit, the traffic implications and ongoing search behavior down the path to conversion must be considered if leaving the URL out of offline creative planning. This is especially true for large brands with thousands of potential visits per day per category or line of business.
All that said, offline marketing alone, even without the mention of the URL, will affect search. The more that a brand places its products or benefits in front of potential consumers, the more likely that subsequent branded search behavior will occur.
Integrated Segmentation – Reaching the Next Level
It is clear to those involved in large-scale advertising that there is a strong connection between all the components in the marketing mix. The next step is to ensure that each component speaks the same language to the same target markets. Just as they customize creative based on the marketing channel and likely target personas, advertisers have to think through the cycle. Having one voice or look and feel within ODA or offline marketing, but a completely different brand experience at the web, can lead to confusion and less trust.
Once marketers begin to understand exactly how their target consumers interact with different forms of advertising, and more insight into the paths typically followed to conversion, the big picture needs assessment to make sure everything flows smoothly and expectedly. This will lead to greater collective ROI, and continuously growing data samples to further refine strategies from.