The biggest search engine marketing conference of Q1 is behind us. Let's review the three biggest PPC trends and insights from SMX West 2017 on attribution, targeting and local search. Join our webinar for even more actionable local search tactics.
(pictured: Bart Heideman of @google discussing the publisher's attribution models.)
SMX West 2017 - the Q1 trade show that unites search engine marketing professionals with prospective clients and search publishers in Google’s backyard - is behind us. What were the biggest paid search insights and takeaways of the show? Let’s take a look now:
1.) Attribution: Last-click versus Top-Funnel / Mobile versus Everything Else
If you work in SEM, you know this challenge well. As we continue to see the online customer journey transmute into something very different than a traditional sales funnel, with prospects jumping back and forth between different stages and across different channels and devices, it becomes harder and harder to clearly attribute one step or another with the one clear conversion that led to the sale.
All parties - Google included - appear to believe that even though there is still no neat-and-tidy solution to the challenge of cross-device attribution - that top-funnel needs to be emphasized further. While it remains challenging to directly translate top-funnel metrics, like the number of impressions a display ad might get, to direct conversions or final sale numbers, all parties seem to agree that top-funnel marketing must be emphasized further.
Google’s position remains the same: The publisher recommends de-emphasizing “last-click attribution” and, if at all possible, switching to data-driven attribution, which utilizes Google’s proprietary algorithms to calculate attribution to various segments of the customer journey (and which also requires a minimum threshold of 20,000 clicks and 800 conversions within a 30-day window). Last-click attribution is, of course, the practice of assigning all “credit” for a conversion or sale to the absolute last click in a customer journey preceding that final transaction...and, as argued by both Google and other SEM session speakers, overemphasizes the final step in that journey while completely ignoring the increasingly complicated steps that led to the sale.
In terms of device-based attribution - if you hadn’t heard, 2016 was the year of mobile. After all, in 2016, mobile clicks exceeded desktop clicks for the first time. Which means that 2017 is also the year of mobile (and presumably, 2018 will be as well). There are, as you no doubt know, methods of cross-device conversions through Google Analytics, including reporting overall percentage of website traffic by device through audience reports by using the Audience -> Desktop/Mobile/Tablet view, as well as using the Path Length metric in Google Analytics Conversions to get a sense of the size and shape of your multi-channel conversion paths (be they through paid search, organic, direct or otherwise).
By checking individual paths for keywords and landing page URLs, it’s possible to get a sense of how your mobile campaigns perform against your pure PPC campaigns. In any case, all signs point to increasing importance of higher-funnel marketing activities rather than simply focusing on the final click.
How will Google Assistant change the way we do local search...and search in general?
2.) Updates for Local Search - Maps Ads, Google My Business and Google Assistant
Local search is big, with “near me” searches doubling in volume in 2016. Mobile is a big part of local search, with 83% of “near me” queries performed on mobile. However, there are still potential blue oceans for local, as well as big question marks within the larger Google ecosystem.
Consider Google Maps ads, which surface ads directly in the Maps interface. While there are only limited spaces available, not every business - not even all Fortune 500 companies - are buying out their “local listings.” Google Maps ads are driven by your Google My Business listing - 41% of online users use Google Maps worldwide, and 30% of all Google searches reportedly have local intent. There exist real opportunities to lift local search campaign performance with aggressive updates to your Google My Business page, including time-segmented bidding strategies that bid up during high-traffic hours (lunch hours for a local restaurant, for instance) as well as aggressive updates to ensure your storefronts have updated call numbers, location and hours information and especially local inventory.
(For more specific tactics you can implement immediately, please check out our webinar “6 Best Practices for Local Search to Use Today.”)
In addition to short-term local search tactics, Google is looking to change the game long-term with Google Assistant, its AI-powered software that is already available for Android phones and in various forms for TV and automotive. Assistant uses “Actions” rather than queries, perhaps changing the nature of search from a question-and-answer-based framework to a more-conversational one.
In search, one might use the query “best restaurants near me 94105?” Using Assistant, a user might instead input “Find me an italian restaurant 94105.” Jason Douglas of Google admits that Assistant is still very much “early days” and that numerous linguistic question marks in the query-versus-Actions paradigm have yet to be resolved. When instructed by a user to “Find a pizza place near me,” should Assistant rank results based on proximity or a user’s personal preference to specific types of pizza?
How will factors such as user ratings, price range and proximity affect inquiries for urgent service calls, such as repairing a computer that suddenly broke down...should Assistant locate and call the closest relevant repair outfit even if it has only two stars out of five on Yelp, or should it locate the highest-ranked repair location regardless of price? Should Assistant become “the next big ecosystem for Google” as Douglas suggests, one area of search engine marketing that will clearly be affected, for better or worse, is local.
Are you making use of YouTube’s powerful and varied targeting options?
3. Audience Targeting - Best Practices and New Tactics
With the constant changes in the customer journey, it’s arguably more important than ever to properly target your prospects. There are several tactics that marketers can use to fine-tune their outreach, including tools that were originally earmarked for other purposes.
Gmail Customer Match, which sends promoted emails through the Gmail accounts of prospects, can be used as a remarketing conquest tool by selecting prospects that are receiving messages from your competitors - and proactive marketers can go one step further by also including the search string “unsubscribe me,” “unsatisfactory,” and other terms suggesting a negative experience from customers looking for an alternative service.
Other powerful audience targeting tactics include using the extremely deep and granular targeting options of YouTube, which let marketers target prospects by specific videos watched/liked/commented on/added to a playlist, specific channels to which prospects have subscribed and many others. YouTube reaches more 18-49 year-olds in America than any American cable network with 1 billion active users at an average session time of 40 minutes.
From a strategic perspective, excluding prospects can be just as important as including them, whether prospects be already-known (and already-counted) names deeper in the funnel / show poor engagement / simply aren’t a fit for your company’s desired persona. In order to facilitate testing and optimization, ensure the right message reaches the right audience and avoid bombarding prospects with oversaturation, it can be a good idea to divvy up prospects into separate segments, such as your known remarketing list, audiences sorted by industry, audiences sorted by job title, and the most general audiences that you might pursue with standard tactics like lookalike searches - searches for prospects with similar demographics to your known clients/deep-funnel prospects. For instance, remarketing lists (RLSA) serve a dual purpose as powerful exclusion lists for top-funnel outreach - why waste spend on top-funnel channels such as display trying to drive awareness to prospects who are already in your database?
For more in-depth session coverage of SMX West 2017, please check our previous coverage:
- SMX West 2017: Day 1 - Google Assistant, Advertising in Google Maps, Offline Attribution
- SMX West 2017: Day 2 - AdWords Attribution, Drafts & Experiments, and More Attribution
- SMX West 2017: Day 3 - Audience Targeting, PPC for Brands, Testing and Optimization
- SMX West 2017 Wrap-Up - Top 3 PPC Takeaways
About the Author
Andrew Park is a content marketing manager at QuanticMind. A UC Berkeley graduate and lifelong Bay Area resident, Andrew has done tours of duty in editorial, PR and marketing, and now works with the QuanticMind team to communicate the importance of data science and machine learning in digital advertising.More Content by Andrew Park