Are you wasting money on branded search?

Marketers are always looking for effective ways to cut costs without impacting customer acquisition. The truth is that big cost savings are possible in advertising—if you know where to look.

“Half the money I spend on advertising is wasted; the trouble is I don't know which half.” This complaint, made famous by US advertising pioneer John Wanamaker at the turn of the 19th century, still holds true today. Marketers are always looking for effective ways to cut costs without impacting customer acquisition. 

The truth is that big cost savings are possible in advertising—but it's often hard to figure out where these opportunities lie.

One area that frequently comes under scrutiny is branded search, or buying ads for when people specifically search for your company or product name. In 2013, eBay found that branded search had no incremental impact on users' likelihood to make a purchase. More recently, AirBnb shifted heavily from performance marketing (especially search marketing) to brand marketing after pausing spend during the pandemic, which only led to a 5% drop in traffic.

Ultimately, both these companies have strong brands. Brand search is going to be less incremental when your brand is already really well-known and you are the clear leader in your space. The opposite of that is that branded search is going to be more incremental the more competitive your space is and users searching for your brand name could get distracted by other brands.

Every business should strive to keep close tabs on how branded search is working for them. In this article you will learn how executives can quickly determine whether branded search is the right strategy for their business. In the second part, we also suggest a few systematic and scientific ways to test how incremental your branded search is, as a channel.

Is a Branded Search Ad Strategy Right for Your Business?

Marketing Executives can quickly evaluate the appeal of branded search ads for the business by following these three steps:

  1. Analyze your spend: Look at how much you're spending on branded search. This will help you gauge the size of the potential cost savings available. 
  1. Assess branded keyword competitiveness: The simplest way to do this is to search for your company using your branded keywords. As your competitors get more aggressive in terms of bidding on your brand search terms, then it's going to get more incremental for you.

To achieve more detail, use tools like Google AdWords Keyword Planner to check keyword competitiveness. Any user specifically searching for your brand or product is likely to have high intent. If the Keyword Planner's ‘Competition (indexed value)’ metric is above 33, then you should be prepared to bid on your branded keyword to ensure that the first company to appear in the search results is yours - and not that of a direct competitor. 

If the metric falls below 20, there is likely some increased scope to cut spending. You can also use ‘Ad impression share’ (which indicates the percentage of potential ad impressions you win at auction) to gauge this.  

  1. Check organic search results: Look at the number of organic search results on the page for your company name by searching for your company name and seeing how many of the links on the first page lead to your site. A company with a strong presence in organic search results may be able to rely less on paid search ads to drive traffic to its website.

Testing for incrementality

Using the simple three-step framework above, you should be able to get a rough idea of the attractiveness of branded search for your business. Unless you have highly competitive branded keywords and very few highly ranked organic results, we strongly recommend testing for incrementality.

Testing will ultimately solve two questions:

  1. What incremental return do I get from branded search ads
  2. How does this return stack up to how else I could be using this budget

What Incremental Return Do I Get From Branded Search Ads?

You can measure marketing incrementality to determine the additional value that a specific marketing campaign or channel is bringing to your business. For example, if a user didn’t see your ad, and all other factors remained constant, are they less likely to convert - and if so, by how much?

One method is to compare the performance (say, the number of users who reach your campaign conversion goal) of an active treatment group to an inactive control group. By comparing the number of conversions between the two disparate groups, you can gain a much clearer and more accurate understanding of the true incremental value.

Remember, to keep the control group and the treatment group as similar as possible, all other factors must be held constant. Due to constraints with Google Ads and Bing Ads, it's not possible to run randomized controlled trials within these ad platforms - which makes holding all other factors constant difficult. 

There are a couple of ways to run these tests for incrementality.

  1. Geo testing. Divide the market into geographical units, such as states, DMAs, or zip codes, and then select a treatment group (paid search switched on) and a control group (paid search switched off). The trickiest part is identifying a set of control regions that act like the treatment group. 

A popular emerging approach is to use synthetic control groups. For example, if your French customers act similarly to a weighted average of Germany, the UK, and Italy, you might pause branded search spend in France and compare the impact on conversions to the weighted average of the three other countries. 

Use packages like Meta’s GeoLift to help identify these synthetic control groups. 

  1. Regression discontinuity. In cases where geo testing is not possible, you can temporarily switch off paid search altogether. Compare your results over that specific period to determine branded search’s effectiveness. The simplest way to do this is to study web traffic by search source: organic, paid-search-branded, and paid-search-non-branded. 

For the period where you switched off branded search (paid-search-branded traffic will plunge to 0). If branded search is not incremental, then you can expect to see total traffic remain constant as all of the paid-search-branded traffic switches to ‘organic’. 

More sophisticated approaches will control for other factors (such as seasonality) and focus on the impact on conversions, rather than traffic alone. 

These tests will help you understand “if i don’t pay for branded search ads, how many sales will I lose”. We have worked with companies where 80%+ of conversions would have happened if your ad was not present. That’s definitely something worth including in your marketing reports!

How Do the Expected Returns Compare to Alternative Uses For This Budget?

Let’s say you spend $1m per year on branded search ads and achieve 100,000 sign-ups as a result. That means you have a customer acquisition cost of $10. Assuming only 45% of those conversions are incremental then branded search is only actually responsible for 45,000 signups. This pushes your CAC from branded search to ~$22.

In the example above, it may be that Meta ads have an incremental CAC of $15 and therefore you can acquire the 45,000 incremental signups for $675,000 instead of $1m with branded search, saving you a whopping $325,000 whilst still hitting your signup numbers.*

*(Illustrative example)


Ultimately, the goal is to figure out if branded search is a worthwhile undertaking for your business - after having adjusted for incremental sales only. By evaluating the framework and testing for incrementality, you can make an informed decision about whether to continue investing in branded search or shift your focus to other channels. It’s important to note that incrementality of branded search varies over time so it is important to continually test and monitor performance.

Ready to unlock new
growth opportunities?

We and selected third parties collect personal information. You can provide or deny-  your consent to the processing of your sensitive personal information at any time via the “Accept” and “Reject” buttons.