The rise of automation: Performance Marketer v Machine

Automation and AI technology have been leveraged in marketing platforms for years and that is only set to mature and evolve. The ‘threat’ of automation to the traditional value of paid media specialists has been evident for some time and with that has come the gradual evolution of the role and function of digital marketing teams.

Since the 2010s, Google has supercharged automation, with one common refrain from those working in the industry “How is this going to make Google more money?” From the perspective of a pay per click marketer who has been perfecting their craft for years by manually optimising accounts, it’s always viewed as having less control. 

Last year Google launched Performance Max – a campaign type that takes creative, audience signals and conversion targets and serves them across all of their platforms including YouTube, Gmail, display and search. While still early days relatively speaking, this new Google landgrab seems to work and can’t be ignored.

The keyword here is search. Serving ads based on signals outside of keywords is not new, as dynamic search ads (DSAs) have been doing this for a long time now. What’s new is that Google will serve ads across their full ecosystem, whilst not giving any information about where those ads have been served (save basic display placement reports). This includes on search, and they certainly don’t give you search term reports to analyse. In this light, Google is for the first time treating search as another platform it can serve on to achieve results.

Performance marketers have to make a decision: continue fighting the machine, or recognise it for what it is and develop strategies that work alongside the machine.

It is not the end of traditional search campaigns, not yet. Google will still prioritise keywords that match queries inside of traditional search campaigns, and it’s important to continue thinking carefully about account structure.

The shift in importance comes around creative, audience and data. Performance marketers need to move away from intricate micro optimisations and move to an audience first approach with data and creative key strategic considerations. The value is how we make better use of data, target the right people, and serve impactful ads.

Who are the right people to get in front of? How do we plug the best data in to create real impact? What will they care about seeing?

Paid media automation is only as good as the data that’s fed into the machine. If just the first part of a conversion journey, such a lead form submission, is plugged in, automation will not take into account the quality of that lead. Using the wrong conversion point can therefore result in automation driving results that are not aligned to business objectives or outcomes.

The web of softwares and technologies will be a hurdle for some marketers, or a barrier to others. Spending time capturing and feeding the touch points as far down the conversion funnel as possible to fully take advantage of automation is required. Strategically, integrating and analysing broader data sets improves the entire marketing effort, enabling businesses to measure true marketing effectiveness.

It’s common to disregard the importance of creative in marketing, particularly in digital marketing. The end user will not appreciate the cutting edge technology used to serve an ad at the perfect time, but they will care about what message is being conveyed. This again needs careful consideration. Creative will play a larger role as we diversify channels and focus less on what we lean on the machine to do. Creative needs to resonate with audiences across channels and from the top to the bottom of the funnel. Leveraging creative thinkers to solve customer problems through creative assets is imperative.

Paid media specialists have to move away from intricate micro optimisations to an audience first approach. Their value lies in making better use of data, targeting the right people and serving truly impactful ads.

The case of performance marketer v machine is not yet resolved, however, a change in approach from the human side of this equation is needed now more than ever.