Here are some of my tips on running and optimizing Google Display campaigns + case study.

As we have discussed in my other post (๐˜ช๐˜ง ๐˜บ๐˜ฐ๐˜ถ ๐˜ฉ๐˜ข๐˜ท๐˜ฆ๐˜ฏ’๐˜ต ๐˜ด๐˜ฆ๐˜ฆ๐˜ฏ ๐˜ช๐˜ต, ๐˜ด๐˜ค๐˜ณ๐˜ฐ๐˜ญ๐˜ญ ๐˜ต๐˜ฉ๐˜ณ๐˜ฐ๐˜ถ๐˜จ๐˜ฉ ๐˜ฎ๐˜บ ๐˜ฑ๐˜ณ๐˜ฐ๐˜ง๐˜ช๐˜ญ๐˜ฆ ๐˜ฐ๐˜ณ ๐˜ด๐˜ฆ๐˜ข๐˜ณ๐˜ค๐˜ฉ ๐˜ฎ๐˜บ ๐˜ฑ๐˜ฐ๐˜ด๐˜ต๐˜ด ๐˜ฉ๐˜ฆ๐˜ณ๐˜ฆ ๐˜ช๐˜ฏ ๐˜จ๐˜ณ๐˜ฐ๐˜ถ๐˜ฑ), a good offer is ๐ฆ๐š๐ง๐๐š๐ญ๐จ๐ซ๐ฒ.

If your offer is ok-ish, in the best-case scenario where you have killer ads and on spot targeting, you will get some results, but that won’t turn into a fantastic account.

If you aren’t using Facebook ads for testing visuals, you can start a Display campaign targeting desktop and mobile traffic in separate ad groups. This way, you can get a full grasp of which images and copy work the best. Make sure you allow enough stats before declaring a winner. After the test phase, copy the winners into the Smart Display with Pay per Conversions bidding strategy. With Smart Display, you can’t manage devices, so I suggest only try those ads that proven to work in a standard display campaign. The budget for Smart Display is 3xCPA.

๐“๐ž๐ฌ๐ญ ๐š๐ฌ ๐ฆ๐š๐ง๐ฒ ๐ซ๐ž๐ฌ๐ฉ๐จ๐ง๐ฌ๐ข๐ฏ๐ž ๐š๐๐ฌ.

I have been loading mine to the max:

  • at least 4-5 images (if you have 15 go for it)
  • a logo,
  • 5 headlines,
  • 1 long headline and
  • 5 descriptions

If it sounds too much to you, with ad quality meter in the responsive ad builder, you can see what machine thinks of your ad, so make sure it shows the highest quality before you run the ad.

While we are talking about responsive ads creation, I suggest testing different CTA buttons. For my sales-oriented campaigns, my usual A/B/C CTA test is “Shop Now,” “Learn more,” or automated (Google optimize it).

Responsive ads are not going to give better results vs. static banners for every business. That’s why I suggest testing your responsive ad winners with static ads as the next test in the line. Check spy tools like SEMRush to get an idea of what sort of visuals your competitors are using (๐’๐’๐’• ๐’•๐’ ๐’„๐’๐’‘๐’š ๐’ƒ๐’–๐’• ๐’•๐’ ๐’ˆ๐’†๐’• ๐’‚๐’ ๐’Š๐’…๐’†๐’‚ ๐’‰๐’๐’˜ ๐’•๐’ ๐’Ž๐’‚๐’Œ๐’† ๐’š๐’๐’–๐’“๐’” ๐’…๐’Š๐’‡๐’‡๐’†๐’“๐’†๐’๐’•).

๐ˆ๐ง ๐š๐ง๐ฒ ๐œ๐š๐ฌ๐ž, ๐ฆ๐š๐ค๐ž ๐ฌ๐ฎ๐ซ๐ž ๐ฒ๐จ๐ฎ ๐ก๐š๐ฏ๐ž ๐š๐ -> ๐ฅ๐š๐ง๐๐ข๐ง๐  ๐ฉ๐š๐ ๐ž ๐œ๐จ๐ง๐ ๐ซ๐ฎ๐ž๐ง๐œ๐ฒ!

๐๐ข๐๐๐ข๐ง๐ : if we don’t have any previous running display campaign or we have just set up the account, we will try pay per conversions immediately while setting CPA as high as possible to get the campaign running.

If that won’t start working after a few days, we will go to pay per click. In both cases, our budget will be at least three times our CPA. There is no right or wrong here, I have seen great results with both paying per click and conversions, consider this as an additional experiment you can do in your campaign.

๐“๐š๐ซ๐ ๐ž๐ญ๐ข๐ง๐ : from our cold traffic campaigns, we exclude all visitors that have converted in the last 90 days (from GA).

We find keyword targeting performs much better when we use it as an additional layer.

For placement targeting, spy tool like WhatRunsWhere can give you a hand and let you see all URL your competitors are targeting so can focus on those first. Pretty handy, right.ย 

Topics are one of my favorite targeting options and work the same as interests in Facebook Ads. I suggest you use one topic per ad group until you get enough data. After that, you can play around and layer them with additional targeting.

For affinity and in-audience list targeting, I suggest you spend some time in your Google Analytics and learn more about your current buyers.

Be especially careful with an in-market audience as I find this audience type has the most significant mismatch of what should be doing vs. performing. Due to the kind of eCommerce I run with Google Display, I use the least in-market audience, but that doesn’t mean you won’t see success. That’s why it is essential to test all targeting options.

As we don’t want to make constant changes with our Search campaigns, the same applies to Display.

It would be best if you allowed at least 2-3 weeks before making changes.

Before excluding a targeting option, I give it space up to 3x CPA (after that, I get impatient and just shut it down lol).

Can you make a consistent success with Google Display? If you made it Facebook Ads, you could make it big with Display, as Display is a much bigger platform.

But, if you haven’t had any success with low intent traffic before, you might need to work on your offer and advertising angles before diving into Google Display.

On the PrintScreen below, is displayed account where we kick off with Google Display in May last year.
No previous history, no conversions, blank.

We started working on it with the previous estimate that it’s going to take at about a year to find a good opportunity on Display and ended up with being profitable since month one. Our worst month performs at 200%+ ROAS.

We shorten our learning process leaning on the data our client delivered from their Facebook ads.
The offer, the visuals, different targeting were already tested on Facebook Ads, so logically we started with the best performers first and after that, we started adding new targetings.

So instead of worrying about only one traffic channel, you will get the most impact if you combine sources and lean on the data the business already purchased.

What is your current biggest struggle with this type of campaign?
Let’s discuss Display Network campaigns. Drop a DM on Facebook atย