Over the past two months, I’ve been deep diving into broad match campaigns with Smart Bidding in Google Ads to see how they perform. I’ve always been very cautious when approaching broad match campaigns, but I’ve come to appreciate their effectiveness in expanding reach and driving conversions. Throughout my testing process, I’ve tested different campaign setups, and I’ve gained some valuable insights that I want to share.
Before I get into the details, let me explain why I decided to dive into broad match campaigns in the first place. Earlier this year, I attended a virtual workshop that focused on broad match with Smart Bidding. While I learned a lot from the workshop, I wanted to see for myself if the new broad match was genuinely better. Google provides guidance on their preferred setup, but I was determined to test multiple campaign setups to see which one performs best. Here are the four setups that I tested:
- New campaign with ad groups set up by theme. This campaign type used a portfolio bid strategy, where this campaign and the other campaign with exact and phrase match variants were added.
- New campaign with ad groups set up by theme. This option did NOT use a portfolio bid strategy and was treated independently.
- New campaign set up with ad groups set up by theme. Negatives added with keywords that are already within the account and brand keywords removed.
- Experiment with the broad match keywords added to the existing campaign.
It’s worth noting that when evaluating the success of your broad match campaign, you need to exclude the first week’s worth of data, as the campaign is in learning mode, and there’s not much value in the initial results. As I learned from the workshop, the longer the campaign runs, the better the results get. This is because the campaign continuously learns and adapts to bring you the results you’re trying to achieve.
Now, you’re probably wondering which setup yielded the best results. Well, without a doubt, Option 1 was by far the most effective setup for my broad match campaigns. That is not to say the other options were not great; it was just more clear that Option 1 performed the best.
How did I measure success? I primarily focused on ROAS, as I was focused on getting purchases for the many accounts this was implemented for. After the first month, Option 1 ROAS had an average of 10 ROAS vs. a 5 ROAS for the average across all the broad campaigns. After the second month, efficiency started to drop but was still over the goal with an average of a 6X ROAS. This was also when I noticed other setups improving to between a 3 and a 5 ROAS. I think the biggest learning I gained here is that broad match + Smart Bidding does work – it just requires some patience and ongoing maintenance to see the results you like.
Since I really saw some great success with Option 1, here are the four crucial elements of this setup’s success:
- Use a portfolio bid strategy. Google’s algorithm takes into consideration everything you’ve added, including the exact and phrase variants, when deciding where to show specific keywords.
- Group your keywords by theme. Rather than creating multiple ad groups, condense them to have like keywords in one ad group. For instance, apartments for rent, apartments near me, and apartments for lease, would all go in one ad group. Another ad group can be for 1 bedroom, another for 2 bedrooms, and so on. The system is smart enough to distinguish between them and show more relevant searches.
- Use a relevant URL. This is another signal Google uses when determining which keywords to show. Ideally, you should supply Google with a landing page that matches the search query. For example, if you’re targeting 1-bedroom apartments, then you should supply a landing page for 1-bedroom apartments, and so on. The more specific you are, the more relevant traffic you’ll get.
- Start broad and narrow your goal. Don’t start with the campaign at the tCPA or tROAS that you ultimately want and expect great results, especially if they are very ambitious. Instead, start with a Maximize Conversion or Maximize Conversion Value goal, and after a few weeks to a month, depending on your budget and data gathered, add a tCPA or tROAs goal that matches what the campaign is currently achieving. Continue to change it in 10% increments (more than that tends to tank the campaign) over time, making sure to wait between changes until it comes out of learning mode. Starting with a goal that’s too low for tCPA or too high for tROAS will restrict your campaign to the point where you see no volume and feel that broad match isn’t for you.
I’ve written before about broad match with Smart Bidding, but now I have a better understanding of what it takes to get this type of campaign off the ground. This is definitely not a set-it-and-forget-it campaign type. You need to actively monitor, add negative keywords, and continuously optimize to get the best results from the broad match with Smart Bidding campaign. Give this a chance, and you might be surprised by the results you can achieve!
