Integrating Paid Search into Your Growth Strategy

Paid search has gone through some rapid changes recently, along with similar trends in the real-world application of data-driven channel attribution. Why? Google Ads ‘simplified’ their platform over the years, resulting in reduced control over and insight into your advertising effectiveness that often muddies the attribution waters. That means it’s more difficult to know what’s actually working, and thus, harder to optimize efficiently.

But despite the ever-increasing challenges, paid search can still be a strong pillar of your growth strategy. In this article, we’ll cover effective techniques to take back some of that control and tease out the truth in your metrics, allowing you to drive growth with paid search.

The ‘Broad Matchification of Google Ads’

Google Ads is a constantly evolving platform, and over the years, much of the advanced control marketers previously had has moved into a black box in an effort to promote simplicity. The idea is easy enough to grasp—Google wants everyone to be able to use Google Ads without needing a search engine marketing (SEM) expert to build and optimize every account.

In theory, this sounds great—give people an easy way to leverage paid search as an effective marketing channel. But the reality is, if you don’t know what you’re doing (and sometimes, even if you do) you risk losing control of how your brand is represented in Google Search.

What gets lost in the move to a single ‘on’ switch

The tactics that once worked—even a few months ago—could be losing their effectiveness or be obscured by increasingly blurred channel 'attribution' metrics. To keep things simple in this example, we’ll use 'keyword' to cover key terms and key phrases (multiple words strung together into a phrase) since this is how it’s regarded in Google Ads.

Google has been taking back control of keyword interpretation with the loosening of exact match and the phasing out of phrase match and broad match modifiers. Essentially, Google decides what your keyword means and how to match that with search entries.

That can be great for finding and capitalizing on obscure related terms, relating a search term with your product, or discovering how the customer thinks of your product versus how you market it. But it can also be terrible for restricting or controlling what searches trigger your keywords.

Why the type of keyword matching matters

In the current setup, the context of a keyword can blur into branded terms, competitor terms, and just plain unrelated terms—and you’re paying for all of that. If you don’t have the volume and insight to sort this out, you’ll likely be left in the dark, chasing keyword attribution and untangling account performance when you could be optimizing for increased growth.

Here are some recent Google searches and the broad match keywords that were matched to bid:

If your campaigns have sufficient volume and a conversion-optimizing bid strategy, most of these strange matches will get buried in the vast number of searched queries. Depending on how your account is structured, you may see a great deal of overlap between seemingly unrelated campaigns and ad groups.

To see the effects of matching, you should be monitoring the actual search terms along with what keywords they are matching within the Google Ads interface:

Left Sidebar > Campaigns > Insights & reports > Search terms

Within this report, look for:

• Overlap between unrelated campaigns and ad groups. You may see the same search triggering seemingly unrelated keywords in different campaigns or ad groups.
• Unrelated topics. You may see unrelated searches triggering your keywords; pay particular attention to those with high volume or high CPCs.
• Branded terms in non-branded campaigns. You may see branded terms showing up in a broad match campaign for related services.
• Competitors showing up outside of competitor campaigns. You may see competitor brand names triggering your ads in unexpected places, such as ad groups focused on topical or product feature keywords.

How to fix Broad Search blending

There are a few tactics to fix this Broad Search blending issue. First, you can create multiple exclusion lists for campaigns showing signs of this problem. For example, if you have a branded campaign, you’ll most likely want to restrict brand-triggered terms elsewhere in your account. Create a branded term exclusion list with all branded keywords and apply it to all campaigns other than your branded campaign.

Similarly, you could have a campaign targeting searches that use the word ‘free’ to manage costs from users looking for a free solution. In this situation, you would want to create a ‘free’ keyword exclusion list for all other campaigns.

The Opaqueness of Paid Search Attribution

Along with the shift to broader terms, we’ve also seen a move toward data-driven attribution within Google Ads. Attribution has become increasingly important beyond just a tracked metric and has now become integrated into the automated bidding strategy and Performance Max (PMax) campaigns.

This becomes a problem for marketers and analysts alike when looking at conversion data and metrics like actual ROAS, especially if you compare them with other website analytics platforms. Often you’ll find the channel conversion data does not line up, even across Google Ads and GA4. So how does Google decide which channel to assign a conversion to? It all depends on what type of model you’re working with.

When discussing attribution, as you would with channel conversion attribution, for example, it is imperative that you know what kind of attribution or model is being used to assign credit:

• Last interaction. All the credit for the conversion is attributed to the last channel that the customer interacted with before making the purchase or conversion.
• First interaction. The opposite of last interaction, all the credit for the conversion is attributed to the first channel interaction.
• Last non-direct interaction. A modification of last interaction, this method removes conversions attributed to direct traffic and credits the last non-direct interaction.
• Linear. This model divides the credit for the conversion equally across all channels the user interacted with during their journey.
• Time decay. Attribution credit is given more to the channels that the customer interacted with closer to the time of conversion, with less credit to earlier interactions.
• Position based. A hybrid approach where 40% of the credit is given to the first interaction, 40% is given to the last interaction and the remaining 20% is distributed evenly across the middle interactions.
• Data-driven. This method uses algorithms and machine learning to attribute credit to different channels based on how they contribute to the conversion throughout the customer journey.

Last-click attribution was the gold-star metric when web analytics were first used, likely because it’s the easiest to track. The assumption was that the click that led to a conversion. But we’ve since recognized that most conversions are the culmination of a journey with many other essential steps leading to a decisive action moment. No one click can tell the whole story.

First and last attribution, which largely split the conversion between the discovery event and the conversion event, also had its moment until we incorporated various weighted models, most recently the clear-as-mud data-driven attribution.

Data-driven attribution has been widely embraced—and for good reason. It can track multiple touchpoints and impressions, and include them in the source weighting, leaving you with a clearer picture of the customer journey.

Unfortunately, the algorithms that choose weighting are hidden in the various platforms and provider black boxes. This lack of clarity can lead to, in the best case, a misunderstanding of the data and misrepresentation and manipulation of data in the worst case.

Why the lack of paid search attribution clarity matters

As you introduce and scale your paid search campaigns, you’re likely to see a fantastic return on ad spend (ROAS) metric in Google Ads—particularly when using broad match terms.

Let’s look at an example in which a client sets up a short, low-budget pilot campaign to test the viability of paid search as a revenue-generating channel. The campaign is up and running and the client is excited to see 4x returns with a total ROAS of 317% for the 4-day test. Sounds great and no doubt an SEM professional would be able to scale this into an incredible revenue-generating channel, right? Sure, as long as they don’t have anyone monitoring SEO or channel attribution analytics to show them the dangers of taking simple attribution at face value.

Paid Search Attribution and Organic Cannibalization

The combination of broadening keywords and black box data-driven attribution creates a shell game of attribution, where you’re constantly searching for what is actually driving your results. This often results in marketing budgets following the attribution, moving from SEO to paid search, which players like Google and Meta are eager to encourage within their campaign analytics.

In the example above, the 317% ROAS could actually be boiled down to one thing—ads appearing on the highest-performing organic search engine result pages (SERPs). Typically, most ROI from a generic broad-match campaign comes from branded terms but if you're not a highly recognized brand, you will find that high-performing organic keywords make a major contribution. What this typically looks like in analytics is a drop in conversions from organic and—depending on incrementality—a near-proportional increase in paid search conversions. This is why it's important to keep an eye on organic traffic when you launch paid search, to understand the relationship between the two, and identify when cannibalization is an issue.

So why pay for conversions you will most likely get organically when you can use that budget to drive additional results?

Several studies have determined that you get better performance running ads even when you rank organically for a term. Google famously ran a statistical experiment in 2011 showing that ads drive 89% of incremental traffic. What has changed since then? The SERP looks nothing like it did in 2011, or even 2020.

That’s not to say that those findings are no longer true but with an ever-changing and crowded SERP, your placement may not be able to trigger that cumulative effect that would enhance your organic listing. And there are other situations where you may not want to spend the money. The key is to understand where the cannibalization is happening and the incremental gain for that traffic so you aren’t paying more for the same conversions.

Where you might see cannibalization

The most common places you’ll see paid search displacing SEO attribution are:

• Branded terms. You should be ranking in the top spot for most, if not all, of your branded terms—your company name, product names, and trademarked terms.
• Top-ranking organic terms. This could include your category, product features, or terms related to the major topics covered on your website. Keep an eye out for both short and long-tail keywords, but focus on high-cost and high-volume terms as those will eat budget.
• SERP without top-of-page ad placements. Google has several different placements throughout the SERP, many of which lie below the fold or fail to provide the collective influence when matched with your organic result.
• SERP-dominated pages. Results where you or your brand own significant real estate—where your brand or websites occupy the majority of the search results.

Tip: Look at Search Top IS in your Google Ads keywords metrics to see how often a keyword is appearing above organic results.

How to understand true channel performance

The best way to capture actual channel performance is to look at channel performance in aggregate as your baseline and observe changes between channels. Monitor KPI and attribution lift/fall, particularly between organic and paid search sources. Paid search should be responsible for an increase in conversions with a minimal impact on organic, driving overall lift through new or incremental gains.

This is more straightforward if you’re introducing paid search as a new channel since you should observe an overall lift across conversions by the amount delivered by paid search. For example, if average daily conversions are 100 before paid search, and you’re receiving 20 conversions a day attributed to paid search, your overall conversions should be up to 120 per day.

If you’re seeing attribution within Google Ads but not an increase across your overall analytics, it could be because Google Ads is stealing those conversions from your existing channels, with the most likely culprit being organic search.

Assuming you have a paid search account, you can understand these metrics by running tests in Google Ads, particularly with campaigns that use overlapping keywords. Now, you need a method to run an effective test.

In short, there will be cannibalization, but you will need to determine if the cannibalization is overall net-positive. That is, when showing an ad on top of organic results provides a multiplying effect versus when it's wasted. This will play out in conversions and revenue for each major keyword you observe while you try to find the maximum conversion gain at the lowest CAC. You can achieve this by slowly increasing spend for each keyword to find the optimal balance.

How to fix the attribution issue

By restricting and excluding organic keywords in your paid campaigns, you can restore attribution to organic channels. If nothing else, you will regain control over how and when ads are shown for your top-ranking organic keywords. Here’s what you should be paying attention to:

• Branded terms. If you don’t already, you should have a campaign dedicated to branded terms. Create an exclusion list that covers the entire account except for your branded terms campaign and add all brand keyword variations that show up in the Search Term Insights report. Check weekly for new branded terms anywhere broad match is used and add them to the exclusion list.
• Top-ranking organic keywords. Look at your top-ranking organic keywords through Google Search Console or your favorite SEO monitoring tool. Find the keywords where you consistently rank in the top spot over time. Pay specific attention to those keywords that are either high-volume or high-cost, as these are the ones that will have the greatest impact on your paid search budgets.

After determining which keywords would provide an overall benefit by being excluded, create a new global exclusion list with these exact match keywords. Monitor and update frequently as the rankings change. To make life easier, create an API script to automate this process daily. We recommend exploring an automated solution for big sites that have a large number of high-ranked organic keywords.

Maximizing Efficient Spend by Reframing Attribution

With the loss of control and insights within Google Ads and increasingly misaligned attribution, how can marketers mitigate these challenges when developing their paid search strategy?

If you’re using any Google product out of the box for metrics (like Google Ads, PMax, GA4), you will not see the whole picture of your attribution. Even with tweaking, the full story remains blurry, partly because it’s in Google’s best interest, but also because attribution isn’t entirely straightforward.

For many decisions, much of what matters happens offline or in places that aren’t specifically tracked and monitored (such as chat, social, email, offline, word-of-mouth, and our brains). Reintroducing yourself to pre-tracking and print-era vanity metrics like lift, sentiment, mentions, reach, and impressions can help tell the story of your conversion journeys.

Generally, for strategic and high-value keywords, you want to have paid ads showing up when:

• Competitor ads are showing up in the SERP
• You do not rank in the top organic position
• You have a poor SERP presence

Testing to Understand the Overlap

To fully understand how paid search and organic keywords overlap, and to tap into maximum performance for both, you should implement an impact test for your branded terms. This simple test will saturate your brand exposure and give you a better idea of the overlap and potential incremental gains from ads but does not factor in variability from seasonality, traffic trends, or other factors.

For this test, determine a reasonable budget and CPC to maximize the impression share and volume of the campaign that is sustainable to run for a few weeks or months to eliminate as much variability as possible and recognize trends.

1. Create a Branded Term campaign using your brand's keywords.
2. Set your bidding strategy to target impression share within a reasonable CPC.
3. Maximize budget within your testing limit to drive volume.
4. Compare any drop in organic traffic/conversions from this campaign; if you see a large increase without much change in your organic acquisition/conversion metrics, think about implementing a more permanent branded campaign in your paid search strategy.
5. Run a cost analysis to see if the added cost is offset by an increase in conversion value or KPI and where the optimal conversion threshold lies to find the right balance for your strategy.

You can implement similar tests with high-volume organic keywords and may find with an increasingly crowded SERP, there is often room to grow.

Although attribution is more complex than ever, paid search can still serve as a valuable pillar in your overall growth strategy. By understanding the factors that make attribution less-than-clear, you can decide which approach is right to take back control of your metrics.

Need help integrating paid search into your growth strategy? Contact the experts at Right Side Up at growth@rightsideup.co to get started.