Platforms Series: Leverage Quora’s Targeting to Boost ROI
Published: May 2, 2018
Author: Scott Trevellyan
For today’s post in our “Platforms Week,” Digital Marketing Specialist Scott Trevallyan explains Quora’s various audience targeting capabilities.
It’s not hard to understand why companies and advertisers are turning to newer channels like Quora to bring value and efficiency to their advertising efforts. Between high intent, cheaper CPCs, and a rapidly growing user base that is carefully categorized within the platform, what’s not to like?
Like any newer platform, Quora has continued to develop its functionality. As a follow-up to a previous article more generally about getting started on Quora, here, I will go into a bit more depth about Quora’s evolved targeting options.
Topic targeting is contextually aligned; you can enter a topic and get a list of suggestions based upon the existing topics with in the platform. All questions posted by users are organized by and tagged with all applicable topics. Thus ads will only show in question feeds that are tagged with the topic(s) you select.
You can combine as many topics as you’d like, but unfortunately, at this point you cannot view the performance of individual topics. For this reason, we recommend trying to keep them as close in subject matter as possible so that you can have a better idea of what’s working best. To view the potential weekly impressions for a topic or group of topics, look to the upper right-hand side of the ad group settings:
As you add topics, you will hopefully see the “potential weekly impressions” grow with each topic, which you can see in the summary window (upper-left corner of the ad set settings view. When developing ad sets, there isn’t a hard minimum or maximum number of potential impressions to strive for; however, you will want to ensure that there is a substantive amount (at least 10k or so).
A pro tip here provided by one of our Quora reps is to only enter one topic at a time. While it is more time-consuming than pasting a list of keywords into the topic selector (ad set level), inputting one keyword at time will typically yield more longer-tail topics, providing cheaper CPCs and further increasing potential impressions.
Question targeting is, wait for it…targeting individual questions. Like targeting topics, question targeting is contextually aligned to the content a user is viewing a point in time, though more granular. Though it does take more time to sift through questions as opposed to topics, according to Quora, engagement is very strong because of the increased granularity and relevance. This could be comparable to exact match versus broad match in search. However, keep in mind that this level of specificity does significantly reduce scale.
Unlike targeting by topics and questions, interest targeting is behaviorally aligned. Rather than ads appearing because of the nature of content, ads will be triggered by the nature of users. In other words, ads will show to people with a demonstrated interest in topics, though they might be consuming unrelated content at the time.
Website Traffic: This is Quora’s version of traditional remarketing; it’s based on a pixel that’s placed across your site either manually or through your preferred tag management system. You can target all website visitors, or you can set up URL rules, such as users that have visited URL X but not Y. You can also set up custom rules such as visitors that have visited URLs A, B, and C URLs, but not D or E.
Lookalike Targeting: Quora also offers a lookalike audience option. Similar to lookalike audiences in other channels, you simply select a source audience from which to build a lookalike, and decide upon the target size (1%-10%).
List Match: In March, Quora added a 3rd option for audiences, where you now have the ability to upload email lists to target or exclude. A few uses for this feature include targeting existing leads, excluding existing customers from your ads, and building new lookalike audiences to expand your reach to qualified prospects.
Geographic targeting options include: country, state, city, and zip code. The implementation of this setting will really depend upon individual account goals and history. These breakdowns are obviously helpful if your business goals are geographically specific. However, one important thing to keep in mind here is that you will only be able to measure geographic performance based upon how you break out ad sets. For example, you wouldn’t be able to target the United States and see how individual states/cities/zip codes perform.
Some of these targeting breakouts such as by location can be manual, so take insights from some of the other channels associated with the account and decide to what level it makes sense to break out locations.
For device, you have the choice between desktop and mobile. We highly recommend splitting out desktop and mobile into different ad sets given that the inventory and performance will likely vary. As with geographic options, you will only be able to see how each device performs by breaking them out separately. Fortunately, as you create a new ad set, you can copy settings from an existing ad set.
Currently, you can exclude locations, questions, and audiences (built from URL custom rules). A best practice is to brainstorm and search for any placements that are offensive, controversial, or that could in any way be damaging to the brand. This process can be tedious, but is absolutely worthwhile to ensure brand protection.
When to Use Contextual vs. Behavioral Targeting?
It will vary on an account-by-account basis, and it will depend on the topic and interest inventory, but in most cases I recommend testing both. If they both work, you will maximize your reach. If one performs much better than the other, you can prioritize your budget accordingly. To ensure test validity, try to mirror the list of topics and interests.
Many marketers will wonder, “Will the different targeting option compete and drive up bids?” Fortunately, they will not as they are competing in separate auctions.