Why To Avoid Using Broad Keywords On Google & Bing Ads

At the beginning of 2016, we onboarded a client who was spending over $20,000 a month on Google AdWords. They were formerly working with a PPC company on the West Coast.

In the end, we positively affected all of the client's paid search KPI's: clicks, cost, cost-per-click, click-through-rate, conversions, cost-per-acquisition, and conversion rate/ROI.

In February 2017, we were able to achieve 7.38% less spending than the previous February, yet capture 24.37% more clicks.

Interested in how we did it?

The AdWords account had 17 active campaigns and around 180 active adgroups. It was a nightmare to analyze results or optimize the campaign. Usually a large amount of ad groups are only suitable for companies with multiple products or services (like an e-commerce store), not for a company with one primary service like this client.

Each of these ad groups represented a separate keyword. The main keyword in each adgroup was a broad match, so analyzing and A/B testing the campaign results was impossible.

We'll try to explain what exactly this means. Let's say we're selling the service of "car repair", but we have different ways of searching for this service. For example, people might search for "car repair", "fix my car", or "auto mechanic". We want to know if people searching for "auto mechanic" are more likely to convert compared to people searching for "fix my car". If we know that one of them usually gives a higher conversion rate (more sales), we will probably want to increase the budget for that keyword and decrease the budget for the keywords not performing as well.

So, because of the broad keywords employed in the current strategy, the same search term was triggered by any of the ad groups, as you can see here:

Search Term Keyword (broad match) Ad Group
Car repair Fix my car Fix my car
Car repair Car repair Car repair
Car repair Auto mechanic Auto mechanic

Because of this, we couldn't tell exactly how well the keyword is performing, if we can't see all of the information in one place. We can't be sure about bid changes or keyword performance.

Our search terms report for that kind of ad group looks like this:

Ad Group Keyword Search Term
Auto mechanic Auto mechanic Fix my car
Auto mechanic Auto mechanic Car repair
Auto mechanic Auto mechanic Auto mechanic

We needed to change the structure of every campaign to have a better idea about each keyword's performance.

So, we merged ad groups which had the same theme (meaning that they triggered the same search terms).

We went through all search terms that had a conversion within the last 6 months (the good thing about that campaign was that it already had all possible variations of keywords and statistics: we didn't need many more keyword or bid research).

We also improved and tested more types of ad text, updating extensions, adding more negative keywords, and soon the results started showing. We started work in the middle of Q1 2016: for search campaigns on Google the result were great.

Period CPC CTR ROI
Q3/Q4 2015 $1.66 2.48% 72.7%
Q1/Q2 2016 $1.56 5.22% 101%
Q3/Q4 2016 $1.32 6.34% 120.7%s
Q1 2017 $1.25 7.94% 140.7%

In February 2017, we showed 7.38% less spending than the previous February, yet showed 24.37% more clicks.

Google & Bing: Broad Keywords

The results we saw on Bing Ads for rebuilding the campaign in the same manner:

Period CPC CTR ROI
Q4 2015 $2.70 1.7% 50.3%
Q1/Q2 2016 $1.72 1.47% 161.62%
Q3/Q4 2016 $1.08 2.85% 157.2%
Q1 2017 $1.07 9.18% 250.6%

In the end, we positively affected all of the client's paid search KPI's: clicks, cost, cost-per-click, click-through-rate, conversions, cost-per-acquisition, and conversion rate/ROI.

Learn More About AdWords