BACKGROUND Ebay Research Labs recently released a report, entitled Consumer Heterogeneity and Paid Search Effectiveness: A Large Scale Field Experiment, which analyzed the effect of pausing brand keywords in paid search. Ebay found that, without paid search ads for brand keywords, searchers simply migrated to eBay’s organic listings. The report noted, “Shutting paid search advertisements closed one (costly) path to a firm’s website but diverted traffic to the next easiest path (natural search), which is free to the advertiser.” At Performics, we regularly conduct similar experiments for our clients’ brand keywords, and we’ve seen varying results. This Briefing highlights our “test and learn” approach to brand-keyword bidding. It also outlines concerns around pausing brand keywords, including protecting your brand and preserving search visibility. TEST & LEARN By understanding the relationship between paid and organic search, brands can maximize search sales/leads at the lowest allowable cost. At Performics, we call this holistic approach “OneSearch.” OneSearch is an iterative process that requires ongoing testing and optimization per market and keyword. We leverage analytics for methodical experimentation to determine how paid and organic campaigns can work together to impact the bottom line, which includes potentially pausing brand keywords. Our philosophy is to test everything, including lowering paid search visibility if necessary to generate an overall (paid + organic) search ROI. As the below tests show, findings vary widely based on factors like vertical, location, level of competition, device and customer loyalty. While advertisers can learn from eBay’s experiment, the bottom line is that you must test for yourself. Test One: Pausing Brand Keywords for a Technology Company Performics Strategy & Analytics team sought to identify the optimal level of brand-keyword investment to maximize order volume for a technology client. To do this, we had to determine the incremental impact of decreasing brand-keyword investment. Like eBay, the client had strong organic coverage for brand terms; therefore we tested the extent to which organic search could compensate for order loss when paid search brand-keyword spend was set at different levels:
- Brand Keyword Pause: For three weeks, we tested variable brand keyword spend across markets in the Midwest (five regions each with a different percent budget cut) to determine marginal value of brand-keyword bidding
Findings In summary, the client saw a 51 percent drop in total (paid + organic) orders by pausing brand keywords. Even when brand keyword spend was reduced to 60–80 percent of original spend, total orders fell 20 percent. Organic search did not make up for orders lost when paid search brand-keyword budgets were reduced: Test Two: Pausing Brand Keywords for a Retailer There may also be situations when it’s more efficient to pause brand keywords that rank highly in organic search. For instance, another Performics client—with top organic rankings for brand keywords—also sought to determine if bidding on brand in paid search was worth it. Again, we conducted a OneSearch test:
- Brand Keyword Pause: For six weeks in test metros, we paused brand keywords, while keeping generic keywords live. We reinvested the brand-keyword budget into generics.
Findings Like eBay, brand keyword traffic migrated to organic search, while the larger investment in generic keywords drove incremental traffic and revenue:
- 2% increase in total (paid and organic) traffic
- 3.4% increase in total search revenue
Based on this test, we no longer bid on brand keywords for the retailer. When turning off brand keywords, we saw the following results (July 2011 to December 2011):
- 13% increase in total (paid and organic) search traffic
- 30% increase in total search revenue
However, note that the traffic and sales increases occurred because of the increased investment in generic keywords. Every advertiser who considers pausing brand keywords should leverage analytics to discover where they can reinvest those dollars—in generic terms, in Facebook, in display, in website optimization—to drive holistic performance. As our tests show, pausing brand keywords works for some, but not all, advertisers. Additionally, whether to bid on brand isn’t an all or nothing proposition; the best approach is to employ analytics to determine exactly what percentage of brand keywords to pause. Every advertiser is unique—test for your brand! Also, keep in mind that eBay (a search engine itself) is a particularly unique paid search advertiser. Ebay’s findings aren’t surprising given its market power, organic search dominance, customer loyalty and lack of competition on brand terms. Searchers understand what’s available on eBay (they don’t need a brand-keyword paid search ad to tell them). Ebay relies heavily on loyal customers who already have a product demand in mind and are searching to fulfill a specifid purchase desire. There are not many sites outside of eBay, Google, Wikipedia, Craigslist, YouTube, and Amazon that have such a clearly defined and functional use for customers. BRAND PROTECTION Advertisers should also consider bidding on brand in order to protect their trademarks. Many advertisers—including competitors, affiliates, resellers, complementary product sellers, critics and counterfeiters—bid on others’ marks (called “conquesting”) to engage searchers. For instance, brand A may bid on its competitor’s trademark; brand A may employ compelling ad copy (like 30% off) to divert people searching for its competitor. Conquesting can create challenges for your brand like:
- Enabling others to “free ride” on search demand you’ve created for your brand
- Giving competitors the opportunity to capitalize on your negative publicity (i.e. Blockbuster bidding on “Netflix” during Netflix’s pricing/product changes)
- Allowing rogue affiliates to bid above you on brand, thus stealing leads you could have driven yourself
- Enabling counterfeiters to gain visibility by bidding on your marks
Even if you’re not bidding on brand, you should monitor your brand keywords in real-time. This requires trademark monitoring tools—like Performics’ Benchtools—that identify when others are bidding on your trademarks or using your marks in ad copy and URLs. Your trademark monitoring tool should integrate with workflow automation technology to alert your agency’s account person when a conquestor is identified. If the conquestor is violating the search engine’s rules, your account person can contact the search engine to remove the listing. Additionally, your workflow automation technology should ensure that your ad is running above the conquestor. This may involve increasing bid, and then easing back once the conquestor is gone to maintain efficiency. Additionally, search engines and search management platforms are testing new bid rule capabilities that enable advertisers to multiply bids if their competitors suddenly shoot above them in paid search. ORGANIC VISIBILITY Organic visibility for brand keywords is a major factor in determining whether to bid on brand in paid search. However, organic algorithms are constantly changing. Advertisers should implement technology, like Benchtools, to monitor paid and organic positions. Automation allows you to react to organic volatility in real-time by boosting paid search positions accordingly. Additionally, the face of the search results page (SERP) is continually in flux. Most significantly, Google is working on perfecting a series of “apps” within the search page for vertical queries. We’re calling this trend Search App-ification. In Search App-ification, the “apps” push down traditional results. In the below example, the Google Shopping Ads (an in-SERP “app”) draw searchers’ eyes from the top-ranked Amazon organic listing: If you’re a retailer trying to make it solely in traditional SEO, what do you do? Because traditional organic results are increasingly losing real estate, you must insure your presence in these continually evolving “apps.” Or, you’ll have to trump the apps with paid search ads. The point is that organic search visibility isn’t what it used to be; brands that reduce paid search investment (i.e. pausing brand plus keywords) must be cognizant of how the SERP is evolving in their verticals. In summary, whether to bid on brand keywords is a complex decision that requires particularized experimentation/analysis, as well as continual awareness of your search competition and the evolution of the SERP in your vertical. Testing and learning key. Leverage Performics Strategy & Analytics Team to design a test specific to your brand, thus helping you determine the most effective and efficient way to invest each and every dollar online.