RevelOne’s Spotlight Series regularly features insights from top experts in our Interim Expert Network. We cover a broad range of topics at the intersection of marketing, growth, and talent. If you’re interested in exploring these topics further and engaging with one of our 200+ executive or mid-level experts, please contact our team at experts@revel-one.com.
Written by Daniel Pietrucha
Trends toward increased automation and machine-learning technology in digital marketing have decreased focus on audience strategies within paid search. With the ascendancy of Google’s Performance Max ad format, handing more control over to the algorithm – whether for bids, creative, audiences, etc. – is increasingly de rigueur. Indeed, PMax doesn’t even allow traditional audience targeting within paid search context (or even in display and video contexts), relying instead on ‘audience signals’, which are superficially similar, but functionally quite different.
But audience targeting can play a meaningful role in driving better paid search performance, specifically in the current climate where profitability has come to the fore. For teams with the right data sets and ability to execute, more sophisticated audience strategies are worthy of exploration. Before proceeding to a high-level overview of some of these more advanced audience targeting strategies, we’ll provide some brief background on audience targeting in Google.
Audience Types: Narrowing the Focus
Google offers several different options for audience targeting within its platform, from demographic targeting to in-market audiences. Not all of these apply to paid search; a few aren’t available to search and shopping campaigns, while others are better leveraged in display and video formats.
From Google:
- Affinity segments: Reach users based on what they’re passionate about and their habits and interests.
- Custom segments: Custom segments help you reach your ideal audience by entering relevant keywords, URLs, and apps.
- Detailed demographics: Reach users based on long-term life facts.
- Life events: Reach users when they’re amid important life milestones.
- In-market: Reach users based on their recent purchase intent.
- Your data segments: Reach users that have interacted with your business.
Our focus below will be on ‘your data segments’; or data gathered through analytics and CRM platforms. These seem to be the most impactful audiences within a paid search context and those most intimately related to important concerns like profitability and efficient customer acquisition.
Observation vs. Targeting
When applying audiences to search or shopping campaigns, observation and targeting are two options. “Observation” allows for audience-specific data to be collected for a given campaign or ad group without narrowing its scope. In observation, ads remain eligible to serve to users outside of explicitly observed audience segments – so long as the user searches a suitably relevant query.
In contrast, “Targeting” an audience segment restricts ad service to that particular segment. Both audience membership and a relevant search query are necessary conditions for an ad to be triggered.
Each option has its utility, but I tend to see “Observation” as a precursor to “Targeting”. While data from observed segments can influence campaign performance, it’s the shift from passive observation to active targeting that generates more interesting results. The latter approach, however, is a more advanced tactic that’s possibly inappropriate for accounts with limited conversion data or smaller audience segments. We’ll get into some of these challenges in a later section.
Why use audiences in search campaigns?
Audience data for search campaigns would be of little interest if it weren’t actionable. In addition to providing valuable insights that can inform broader marketing strategy, there are direct benefits to paid search performance to utilizing audience lists in search campaigns.
For advertisers leveraging automated bidding, Google notes that first-party audience segments applied in “Observation” provide a signal that informs Smart Bidding (automated) strategies such as Target CPA and Target ROAS. Within this context, curating audience lists that provide meaningful information to the bid engine becomes critical. If users who read multiple pieces of your blog content are more likely to convert, create that audience segment, and get that data in front of the algorithm so it can bid accordingly. For those advertisers still utilizing manual bidding, bid adjustments (e.g. increase bid by 50%) can be applied to particularly high-performing audience segments to increase ad service to those audiences.
While the data captured through “Observation” can positively influence the performance of algorithmic bid strategies, there are additional benefits for advertisers who shift from this passive approach to actively targeting audience lists via search campaigns (including in Google Shopping).
Ad service can be materially different for audience segments when actively targeted via unique search campaigns versus aggregated with other users in generalized search or shopping campaigns. The main reason for this is that Smart Bidding strategies optimize at the campaign and ad group levels, while within those campaigns and ad groups, there can be substantial variation in performance across product sets, keywords, audiences, etc.
In other words, Google will balance performance between various ad group or campaign subcomponents to deliver on high-level return goals, but under the surface, large disparities can persist. By targeting an audience directly, advertisers can effectively tell Google their desired return goal for a specific audience segment, rather than have its performance balanced against a portfolio of other (very different) segments.
What this ultimately means is more control for advertisers. Active “Targeting” of audiences makes possible advanced strategies that provide advertisers with more control over how they engage with prospective and existing customers within a paid search context. Audience “Observation” remains an important phase one in this process, as the data collected in observation provides a benchmark for measuring the shift from passive to active audience targeting within search.
Below we’ll explore some of these advanced strategies at a high level, with a particular focus on audience lists related to user engagement and customer data.
Advanced Audience Strategies in Paid Search: Select Applications
1. Control customer acquisition costs
a. Acquiring new customers efficiently requires setting the right targets and effectively managing to those targets. Even if you have the math right on CAC, executing with precision will be a challenge when prospective and existing customers are targeted within the same campaign – and with the same efficiency target. Smart Bidding strategies optimize toward specific CPA or ROAS goals at the campaign or ad group level. While the overall bid strategy may be meeting its efficiency target, it could be overspending on new customers and bolstering the campaign average with high-performing buyer segments. The overarching principle of treating like cases alike and different cases differently favors breaking these very different user groups into distinct campaigns where they can be managed to their preferred efficiency. The benefits are transparency, control, and greater profitability.
2. Leverage customer lists to re-engage buyer segments
a. Retain customers by showing up in the right auctions. Customers aren’t always loyal. If they’re searching for a product you sell, but not your specific brand, they’re not necessarily committed to purchasing from you, even if their overall purchase intent is high. Targeting existing customers in non-brand search and shopping campaigns can help advertisers show up at critical moments when a sale might otherwise be lost to a competitor.
b. Manage efficiency to re-engage buyers profitably. Non-brand search and shopping campaigns can generate incremental sales from existing customers. Still, it may not be profitable to drive sales from these users at the same CPA or ROAS targets used in customer acquisition campaigns. Actively targeting buyers in separate campaigns allows for algorithmic bidding strategies to optimize for CPA or ROAS targets unique to buyers that are consistent with bottom-line goals.
c. Turn good customers into great ones. Leverage audience targeting to encourage more first-time purchasers to move into high lifetime value (LTV) customer segments by getting back in front of them during high-purchase intent moments. Maybe you have data that suggests that customers purchasing 2x in the first six months have materially higher LTV than the average customer, or that three purchases in the first year turn an average customer into a brand evangelist. There are many ways to approach this strategy, with the limitations primarily defined by the size of customer data sets and your ability to harvest unique insights that can be capitalized on in marketing contexts; companies like Wilde can help with the latter.
3. Retarget prospective customers near the bottom of the funnel
a. The customer journey is complex. Users often require multiple touches before conversion, especially for high-consideration products that are not only expensive but research-intensive. Musical instruments are a great example. While price points vary, high-end instruments can cost several thousands of dollars. Prospective customers spend hours watching comparison and review videos, reading blog content, and perusing technical details before ultimately making a purchase decision. Keeping a close eye on engagement signals – and bidding aggressively for users in search and shopping auctions when they are near the bottom of the funnel – ensures that investment in video and blog content isn’t going to waste. This approach can be particularly important for service industries where the sales cycle is much longer and the average deal value is much higher.
Brief Notes on PMax
Many advertisers and agencies are going all in on PMax, with Tinuiti reporting that 91% of their advertisers featured PMax campaigns as part of their marketing mix in Q4 2023. Interestingly, some other large agencies have been publicly signaling a shift away from this highly-automated format.
But regardless of one’s position on the importance of Performance Max campaigns within paid search accounts, the reality is that most of the strategies above cannot be executed – at least not with the same precision – when utilizing PMax versus traditional shopping or search campaigns.
As noted above, PMax does not allow for traditional audience targeting. Instead, PMax enables advertisers to provide audience signals, which are analogous to layering in audiences via ‘observation’ mode in search and shopping campaigns. But while analogous, there is – consistent with the general orientation of PMax – far less transparency: advertisers cannot observe key performance metrics (KPIs) for the specific audience applied as a signal.
PMax does have a new customer acquisition capability, so it is possible to segment prospective and existing customers and target them via unique campaigns. However, there are important differences in control over ad placements. The strategies above focus on delivering ads to users at critical junctures within search and shopping campaigns, but PMax doesn’t as yet allow advertisers to determine which ad format incremental ad spend is allocated toward. While your intention may be to be more aggressive within shopping auctions for certain audience segments, the ultimate effect may be a higher frequency of retargeting ads for this group of users.
This isn’t to say that PMax isn’t the right ad format for certain advertisers. Especially for those with lower budgets and limited conversion data, it may be a good first foray into the world of paid search, allowing for multi-ad format exposure focused on driving higher overall conversion volume. But for advertisers with more robust data sets and refined KPIs, there is a material reduction in control that comes with PMax. In some cases the improvement in performance is worth the reduced control and transparency; in many cases, it is not.
Complexities and Challenges
While the strategies above can yield interesting results, they’re also more difficult to execute, both in terms of creating the proper account structure and maintaining that structure. Limitations within particular platforms create additional challenges. Below are a few things to consider before implementing the strategies outlined above.
1. Google’s ability to matchback users isn’t perfect. One of the ways an advertiser might identify existing customers is through CRM data, which can be uploaded to Google to create audience lists composed of those particular customers. However Google cannot correctly identify every user based on this data, so the split between new and existing customers will be imperfect. It’s worth pairing Customer Match lists with GA4 audiences to increase the effectiveness of new versus existing customer segmentation (though this too is imperfect). For those with the right systems in place, new-to-file data should be monitored to gauge the efficacy of this approach to segmentation and calibrate efficiency targets accordingly. Note also that these CRM lists must be periodically updated to ensure data accuracy.
2. Close attention must be paid to account structure. Because of platform limitations on Google, effecting a split between prospective and existing customers requires utilizing campaign priority rules that funnel users to the desired campaign. Priority rules determine which campaign will take priority in the event multiple campaigns are eligible in the same auction. For example, a high-priority campaign might target existing customers; if a user is in the targeted audience segment, they will be served an ad from that campaign; if the user is not in that segment, they will receive an ad from a lower priority campaign. By funneling out existing customers via high-priority campaigns, lower-priority campaigns effectively focus on new customer acquisition. But things can quickly get complicated if there are multiple criteria for segmentation (e.g. product types, profit margin, additional audiences, etc.) Advertisers must evaluate the logic of the campaign structure to ensure it’s set up to flow users as intended.
3. Complex account structures have increased maintenance costs. Once you’ve confirmed the logic of a new account structure, new campaign launches need to be evaluated for how they fit in with the existing ecosystem. Any advertiser doing hands-on-keyboard work must develop a solid understanding of the structure in place so that new campaigns don’t disrupt the existing logic. Complex structures tend to be more fragile; they’re easier to break and harder to fix when they do. Adopting appropriate SOPs for new campaign launches and developing the right monitoring mechanisms can help mitigate some of the risks that complexity creates.
4. Conversion volume is critical. Conversion volume is the lifeblood of algorithmic bidding. An advertiser could develop the perfect campaign logic for their particular goals and adopt procedures that ensure effective execution, but if the conversions aren’t there, then better performance might be achieved through a simpler campaign structure. The targeting strategies above are largely the purview of advertisers with larger budgets and longer account history; for everyone else, leveraging audiences in “Observation” might be the best that can be done for now, though it can lay the groundwork for more complex applications as additional scale is achieved.
Conclusion
Gaining a better understanding of how audience targeting functions within paid search enables advertisers to vet advanced strategies that provide increased control and transparency. While not appropriate to every advertiser, these strategies can offer significant benefits – especially concerning efficiency and profitability – for those with the necessary resources for successful execution.
About the Author
Daniel Pietrucha is the owner of 174 bpm, a performance marketing consultancy focused on paid search advertising. He has over a decade of experience in digital marketing on both the agency and client sides, working for companies like Orvis, The Pro’s Closet, and Pure Hockey, as well as early-stage DTC companies in the e-commerce space.
About RevelOne
RevelOne is a leading go-to-market advisory and recruiting firm. We help hundreds of VC/PE-backed companies each year leverage the right resources to achieve more profitable growth. We do 250+ retained searches a year in Marketing and Sales roles from C-level on down for some of the most recognized names in tech. In addition to our Search Practice, our Interim Expert Network includes 200+ vetted expert contractors – executive-level leaders and head-of/director-level functional experts – available for interim or fractional engagements. For help in any of these areas, contact us.
