The ancient Greek philosopher Heraclitus is perhaps most famous for saying, “The only constant is change.”
While Heraclitus was not a digital marketer, he sure predicted our current marketing state!
In the early days of paid search, marketers had to do everything manually to achieve business goals. They wrote hundreds of ad copy variations to drive qualified traffic while managing thousands of keyword-level bids to minimize the cost of every click and conversion.
Thankfully, much has evolved since then, and advances in automation have expanded what’s possible. The best marketers no longer consider themselves “doers” as machine learning has started to automate and replace more common tasks. 78% of marketers say they feel relieved that giving a larger role to digital tools will free up time for other strategic priorities such as consulting on digital transformation and uncovering key insights to meet larger business goals.
At ROI·DNA, we’ve refocused our saved bandwidth from “doer” tasks toward strategic partnerships that improve client performance, like our involvement in the Google Premier Partner program or our participation in Google’s International Growth Program. We’ve also broken down the silo that typically separates SEO & SEM specialists at other agencies. We share keywords and performance data across the aisle, working together to map out a full-funnel strategy that increases performance and ultimately drives revenue.
The Rise Of Automation
Google has invested heavily in machine learning since the early days of search. Now, automated capabilities assist (or, in some cases, replace) paid search’s most labor-intensive activities. Some examples:
Performance Max is a relatively new offering, allowing marketers to automatically place ads on Google’s owned & operated properties where they’re most likely to convert. As the name suggests, Performance Max maximizes conversions, unlocks new audiences, and targets based on creative. Accomplishing similar results with a more traditional approach would require a marketer to manually manage several campaigns, missing out on the efficiencies gained through recent advances in machine learning.
Bidding capabilities like Target Cost-Per-Action (tCPA) use advanced machine learning to automatically tailor paid search bids to unique searches. With over 40 queries per second on Google alone, it would be impossible for a marketer to manually achieve a similar bid frequency and optimization pace.
Dynamic Search Ads
Dynamic Search Ads allow a marketer to use Google’s machine learning to capture queries not located within the designated keyword list. Google reviews the content on the marketer’s site and will automatically serve ads to searchers it believes are in-market for the site’s products or services. Google will also automatically generate headlines and landing pages appropriate to the searcher’s intent. All the marketer has to do is add a creative description.
Responsive Search Ads
Introduced in 2018 and Google’s default text ad unit since 2022, Responsive Search Ads allow marketers to upload up to fifteen headlines and four description lines. Google will then mix and match the text assets to determine what variation performs best against a specific query. Accomplishing this level of multivariate testing with the older Expanded Text Ads unit would take far longer and require more management to achieve similar results.
The History of Keyword Match Types
Another great example is the evolution of Google’s match types. Match types tell search engines how far a marketer is willing to distribute their ads across search queries. A marketer who uses Broad Match on a keyword is telling the system to allow their ads to display against the keyword’s phrase, similar phrases, singular or plural forms, misspellings, synonyms, stemmings (such as “run” and “running”), and other relevant variations. However, if a marketer determines Broad Match’s reach is too broad, and ads are being served against queries that don’t contribute to their ROI, they can use more restrictive match types like Exact Match.
When Google introduced Exact Match in the early 2000s, it meant the keyword would only place an ad exactly against the searcher’s query. A keyword [e.g., business software] would only serve an ad against that identical query [business software]. However, Google’s definition of “exact” has morphed over time. In 2014, Google allowed close plural and misspelling variants to trigger ads from Exact Match terms, and in 2017, Google allowed different word order and function words to trigger ads. In 2019, Google announced another change to close variants. Exact Match keywords can now match to queries that share the same meaning as the keyword, including implied words and paraphrases as determined through Google’s machine learning capabilities.
As these changes multiply across English keywords, we still maintain an appropriate skepticism that the system won’t be right 100% of the time. We analyze Search Query Reports to determine how well Google is matching Exact Match ads to queries and improve Exact Match ROIs by implementing new negative terms that narrow query targeting.
Understanding the Realities of Automation
Automation isn’t a silver bullet, and marketers should be cautiously optimistic about how far machine learning can propel them. For example, let’s say a marketer is promoting a tax service. Machine learning cannot understand the importance of April 15th, the last day for procrastinators to file taxes. As a result, automation may incorrectly maintain elevated spending levels when conversion behavior is less favorable on April 16th. Despite all the advances in machine learning, it’s still incapable of understanding why something happened.
Before embarking on any endeavor in automation, it’s also essential to ensure your measurement framework is accurate and ideally includes first-party data. Even with access to first-party data, 85% of digital media professionals cited cookie loss and measurement as two of their top challenges in a recent survey. After all, we can only expect our automated models to be as good as the data it’s trained on.
Understanding these limitations, we must anchor everything we do in a measurement framework, meaning measurement is a part of the conversation from day one and not an afterthought. At ROI·DNA, we check performance and settings daily to ensure the algorithms are taking appropriate actions. If we identify underperformance, we place manual overrides to course-correct an action. Not only does manual oversight and intervention keep our campaigns on track, but it also teaches the system the appropriate actions over time.
The Role Of Marketers In The Age Of Automation
Even with these benefits, automation is reshaping – not replacing – the role of the paid search marketer. They must be great media managers to manage assets appropriately, creative to write text ads that resonate with searchers, and data scientists to interpret performance numbers. The role of a paid search marketing expert is constantly evolving, and it’s our job to continue pushing forward to improve performance.
As machine learning further entrenches itself in paid search, a firm understanding of how these algorithms work will be a crucial skill for managing them in a way that maximizes revenue per dollar. Programming can make a paid search marketer stand out, adding unique value in a world where most ads are managed by the same algorithms.
With more advanced bidding algorithms, inputs are of crucial importance. This is where the paid search marketer’s strategic expertise shines. From bidding to value (placing values on higher quality conversions) to Target ROAS, insight both into campaign performance and pipeline data are essential. Additionally, setting up CRM integrations and being able to interpret pipeline data are both necessary to optimize performance. This sort of advanced strategic thinking can only come from a seasoned marketer.
The Future of Paid Search & Automation
While machine learning will continue to expand its capabilities and grow more sophisticated, it’s just one component of the marketing strategy. As any good marketer knows, the consumer should be at the heart of everything, including a search query. Machines certainly have gotten smarter in recent years, but they haven’t made human oversight and strategic guidance obsolete. To learn more about how to successfully blend machine learning and human insight to drive stronger performance, reach out and say hey!