July 23, 2019

Google loves to keep search marketers on their toes. Whether it’s through changing the algorithm that affects how Google presents rankings or adding new ways to target users in the digital space, it’s fair to say the tools and skills that search marketers use change far faster than the seasons.

While we try to remain on top—and ahead—of the ever-changing landscape via in-depth blogs, webinars, and Twitter, there’s nothing quite like gathering a bunch of search nerds in the same room. That’s when we get to talk about what campaigns we’ve run and tested and the challenges we faced to earn results for our clients.

That’s what I got to do at SMX Advanced, the “only conference that’s 100% devoted to elite search marketing training.” The conference brings together search engine marketers in every industry from around the world to share actionable insights we can take home for our own work.

Below are three of my most important takeaways viewed through the lens of higher education.

Machine Learning and PPC

We’re smack dab in the middle of the transition to machine learning taking over key aspects of paid media, such as bidding, audience targeting, ad copywriting, and the like. That doesn’t always mean we can trust it. Ginny Marvin, editor-in-chief of SearchEngineLand, started the conference off by comparing the machine learning algorithm to a child. A kid doesn’t know much in the beginning, but with the right training and learning, they become far more effective.

Frederick Vallaeys, author of the newly published book “Digital Marketing in an AI World: Futureproofing Your PPC Agency,” shared how to strike the balance of allowing the machine learning algorithm into your PPC workflow. His first point piggybacked off what Nicolas Darveau-Garneau, Google’s Chief Search Evangelist, noted during the kickoff keynote. The algorithm does not know anything about your industry or higher ed in particular—all it knows is how to bid, target, etc.

Like any paid media campaign, you’ll go through a break-in period as you optimize the campaign to make it more effective and teach the algorithm what you want. Once the measurement and goal data are correct, the machine can do the math. However, we cannot rely on the machine to understand that information from the start without a little help. The more data you feed the machine, the more efficient it becomes.

Increase the effectiveness of your PPC testing

Each year, the conference hosts one of my favorite sessions, “The Mad Scientists of Paid Search.” All three speakers provided great insights on things they’ve tested over the last year as well as how to conduct efficient tests. Maria Cocoran, worldwide search marketing media manager for Adobe Systems, made several points on her approach to testing that particularly resonated with me.

First, she discussed an extremely common problem for Search Engine Marketers (SEMs):

1. We want to test everything.
2. We get caught up in campaign daily maintenance and fresh keyword and ad copy implementation, leaving little time for testing.
3. A new feature or product launches, and we suddenly have a new vigor…
4. Repeat.

It’s a cyclical process that most SEMs face, and it leads to broad, potentially vague testing, mostly just tracked in our heads with little to no method.

She suggested deconstructing the variables of search: landing page content, ad copy, keyword match type, audience, etc. and write a hypothesis for each test. This allows you to list and rank your potential tests by upside, confidence, and effort. This provides a more holistic view of your test options, what you can gain, how likely you are to succeed, and how much work it will take.

Cocoran provided an example testing matrix, the spreadsheet which you input all of this information into. I’ve started to use it, and it has already accelerated my testing process and made each test more accountable and measurable. Her scientific method-style approach to testing moves you away from gut feeling and backs everything with metrics and reasoning.

Data Privacy and Targeting

Data privacy awareness among the public continues to increase, leading to tech companies, such as Apple, implementing Intelligent Tracking Prevention. This shrinks the life of a cookie (tracking code) to just 24 hours. Shrinking cookie life, paired with platforms such as Facebook, LinkedIn, and Twitter becoming walled gardens of audience data, have drastically decreased our ability to track users cross-device, cross-platform, and in retargeting efforts without an expensive third-party tracking tool’s help.

One way to help mitigate this is by sharing all of your data across platforms manually and utilizing audience tools on that platform to the best of your ability. Any data captured for marketing purposes should be uploaded to any other platform you use to market online for the purpose of creating a lookalike audience or customer match similar audience. This allows you to focus your targeting even more at no additional cost.

An easy way to implement this practice is by housing your user data in one central location for you to upload to each platform manually. While this will not work as effectively as a third-party tracking tool, it’s much cheaper. The cost of using these tools can be a huge barrier for smaller colleges or universities. Your data is your biggest asset when it comes to cost-effective methods of enhancing your targeting.

Tell us what you think

The speed at which search engine optimization and search engine marketing changes keeps our industry exciting. There’s almost never a stagnant day in the SEO/SEM news cycle, but in the midst of your hustle, take a second to implement these tips.

I would highly recommend attending SMX Advanced if you hope to not only learn about the industry’s leading concepts but also take away action steps to improve your campaigns based on learnings from real people who do the same work you do, day in, day out.