Three Questions, Three-ish Solutions: Proving Out Marketing Trends
Many marketers have access to secondary research (or information from someone important) that says a specific audience gets information from a specific channel and therefore you have the opportunity to use said channel to drive traffic to your website.
More than half of the students surveyed (55.2%) indicated the more interested they are in a school, the more they will interact with the school via social media.
Socialize Your Strategies | Use social media to complement traditional search channels. To fuel that engagement, dedicate and fully integrate a team of digital experts to create visually compelling, customized, carefully timed and student-centered messages about the subjects most likely to prompt students to link to your website—such as college majors and costs.
However, being the cautious marketer you are, you want to prove and validate that it’s useful for you, your objectives and your channels before you go all out creating a new batch of content or invest in ramping up a platform strategy.
Not sure how to do that? Here are a few ideas to get you started.
Question 1: Does <audience> actually exist on my version of <platform>?
Rely on the secondary research — don’t think that your audience is so unique that it bucks generational and societal trends. Another option is to run a focused content campaign designed to only appeal to certain members of the <insert audience name here> — use reach to estimate audience size.
Social media aggregate tools such as Crimson Hexagon, Shareablee and Union Metrics can provide some context about your users based on other areas of the Twitterverse that they engage with. Typically this is data in aggregate, highly assumptive and likely based primarily in Twitter data due to their open data hose. Google Analytics has an (relatively new) Audience Insights section to find basic elements about ages, gender and in-market segment. Use it directionally and carefully — it’s often working from a highly sampled amount of data.
Get to know your platform audience tools. Facebook and YouTube have the most rich audience insights on their own platforms. Use Facebook Audience Insights through the Ad Manager or Business Manager tool — you can segment your owned audience in many ways to size up generations, content lovers and other basic demographics. In YouTube, you can see audience demographics in aggregate or video by video.
Caution: Using vendors to provide audience data is possible, but presents risks. Don’t be the next Mark Zuckerberg.
Question 2: Did <audience> engage with the content I posted for them on <platform>?
Review post performance for content designed to appeal to and drill into any demographics available. If you can demographically identify your audience, did they respond? A more manual approach is to review audience comments in your analysis – can you tell by what people are saying who they are? (i.e. “I’m so proud of my son Jimmy – he made the Dean’s list!” = parent)
Target content to specific demographics where possible. Facebook offers the opportunity to do this organically through limited post targeting options. Are your engagement metrics stronger? When you review the demographics on engagement, do they align with your expectations?
Utilize paid content with focused audience targeting to test whether or not content resonates with a specific group. Extrapolate findings to organic strategies.
If you want to understand engagement with content elements on your website, consider usability testing or surveying via platforms like HotJar or TreeJack. Investing in tools like this will help you understand long-term what’s happening and how your changes improve your user journey. A less costly way to get started is to utilize Google’s native tools for Event Tracking and Goal Conversion.
If you’re unsure about how to understand the impact of your promotions by audience, check out this post – a how-to guide for designing an audience-centric campaign structure using Google’s UTM codes.
What version of the content was most appealing to <audience> on <platform>?
If you’ve had a fairly structured and varietal history on your platforms, review historical data and attempt to classify it in certain ways for interpretation. You can code your data by intended audience, number of words, topic, content, etc… to get a baseline understanding of past trends.
Use your owned platform tools to perform split tests. Emails are a great way to test, but you have to have large enough audiences and engagement rates to get a good read. If you are a member of the University community and want to learn more about all the great email testing happening already, check out the Email Community of Practice
To split test on the web, try Google Experiments. It allows for *almost* development-free testing of words, graphic elements, formatting choices, etc… on your web pages. It will require dev support to get up and running. Also… if you break your homepage, don’t tell anyone this was my idea.
Execute a split test using paid platforms. There are a number of ways to do this, but the easiest two are via Facebook and Google paid search ads. Using paid methods allows you to control variables relative to served impressions, timeframes and audience targeting.
Please use statistics to figure out if differences are significant, otherwise you’re operating on noise. Try an easy web tool like Kissmetrics A/B Significance Test to get started.
To gain efficiency in reporting, try the magical combinations available in Data Studio from Google. It allows for many data connections, the simplest of which is the Google Sheet. Recreate that spreadsheet in Google Sheets, link to Data Studio, create visuals and add narrative, slap on a data selector option and provide access to folks that need it.
Have other ideas about tools to prove out marketing trends? Get engaged with the Analytics Community of Practice on Yammer – start a discussion, ask a question, offer an insight. Can’t wait to have you join!