Sometimes as a media buyer you just need someone that tracks data on millions of dollars of ad spend to just tell you what to do. Or at least give you a framework.
Justin Brooke, in his book Retargeting Recipes, does that. I highly recommend, if you’re in the biz or thinking about using remarketing to goose your business profits, that you get the book. It’s excellent, useful, and great for reference.
“The more times someone sees your ad, the more likely that person is to buy from you. And that’s the beauty of retargeting.” --- Justin Brooke
Specific Remarketing Questions: I Wonder What Justin Brooke Suggests?
Which size Google ads should I request from the design team?
There are 20+ size options, but three sizes get 90% of the impressions. So, I’m going to start there. Those sizes are: 300x250, 728x90, and 300x600.
Which networks should I use to retarget?
Justin suggests Twitter, Google Display Network (GDN), Facebook, and Taboola, because they cover about 98% of the Internet. In a recent Medium article about building “genius pixels”, Brooke says you should be doing the work on Facebook, Instagram, Youtube, and Google Ads.
The article was written after Retargeting Recipes was published, so I don’t know if we’re kicking Twitter and Taboola to the curb, or if Instagram and Youtube are added to the mix. These platforms are always evolving, and it’s tough to write a book that isn’t partially obsolete before it hits the shelves.
However, the strategies and tips are really what matter here. Still, if anyone knows Justin, I’d like to know about which channels I should be putting time into these days. I’m currently retargeting on Youtube, Facebook, Google Display Network, and LinkedIn at the day job.
What are the frequency cap and campaign flight length rules of thumb?
Frequency: 3 times a day. Flight length (how many days do you run an ad to a specific audience) 30 days. The flight lengths vary depending on the recipes, but 30 days is standard.
I want to re-engage customers, what’s the best sequence to do this? Well, first segment them into different cohorts (i.e. purchased within the last 30 days or haven’t purchased in 90 days), then put strategic types of ads in front of them in a precise order and time frame.
Justin also gives suggestions for budget allocation and how long you should run ads to particular pixelled audiences.
He even shares the design of an ad that works best for him, his clients, and his students. The instructions include:
Keep branding and border to distinguish your ad from the content on the page, so it’s not considered deceptive. Even if you get a lift from removing either/both, it’s not worth it if you get popped for deceptive advertising practices.
Also, headline font color should be red, black, or blue. And, font type for the ad should be Arial, Helvetica, Droid Sans, Myriad Pro, or Verdana.
These are the types of answers you want an arm’s length away if you’re setting up retargeting. Or, to give to your team member, who’s in charge of ad buying. Or to be a smart buyer of this service done for you.
Four Types of Ads
Here are the ad types with Brooke’s special catchy names.
- The Brander
- The Product Remind
- The Instant Celebrity
- The Fence Jumper
I don’t want to give away the entire book, so I’ll let you get the book to find out what these mean, but they’re ingenious, and very helpful to open up the mind to the different purposes retargeting ads can serve for your business.
Great Recipe for My Day Job
I run remarketing advertising at my day job. Retargeting Recipes will make me better at my job. It’s more targeted to B2C companies, but I will take many of the concepts and fit them into my B2B, long-sales cycle, training product advertising.
In fact, as I was taking notes on the book, I sent emails to my work address reminding myself to try some of this stuff.
One recipe that stood out was the Earnings Multiplier Recipe. For this one Justin referenced the wisdom of old school marketers, who followed the Recency, Frequency, Monetary (RFM) model to serve ads to different types of customers. This is actually a three-part recipe that includes an audience ladder, which breaks down the audience into five different buckets:
- Leads only custom audience
- All customers custom audience
- 1 purchase only custom audience
- Multi-purchase customers custom audience
- Top 20% spenders custom audience
The recipe details what types of ads to send them and cadences for each. Very helpful stuff. You may need to tweak for your business model, but the solid foundation is there.
It also makes you think. What other types of offerings can we develop that will entice the customers in these different buckets? Next level stuff that will provide your customers more value and maximize your revenue.
Blogger Side Gig Recipe
You may think if you don’t have a product, there’s nothing in Retargeting Recipes for you. Wrong! I don’t have a product connected to this site (yet), but Justin has a recipe for me.
The Blog Monetizer
For this recipe you will need to be promoting an offer. Since I don’t have one, I’m going to add an affiliate product to the mix:
- Pixel people on content pages (blog posts, advertorials, podcast notes pages).
- Exclude customers.
- Instant celebrity or product reminder ad from day 1 to 30
- Fence jumper ad from day 31 to 45
- Brander from day 46 on.
- Budget $10-25/day
Once you wrap your mind around what these ad types are, just follow the recipe.
Sorta Terrifying Side Note: Privacy Cataclysm Is Coming
Justin warns that the retargeting gold rush is in jeopardy ... threatened by privacy concerns (legit and/or manufactured), so make hay while the sun shines.
GDPR. Cambridge Analytica. Browsers blocking pixels by default. More users discovering ad blockers. Government initiatives, particularly in Europe and Canada, to hold marketers very accountable to what they do with customers/visitors data. Etc.
Why Facebook Was Really Down in March 2019 (Indulge in a little conspiracy speculation with me or don't)
Some surmise that Facebook was recently down so the powers that be could help them purge their data of evidence they’ve been Big Brothering the U.S. and world population at large ever since they gave Mark Zuckerberg the keys to Facebook. This handoff, of course, was the day after the U.S. Department of Defense’s Defense Advanced Research Projects Agency (DARPA) “shut down” their creepy data collection project LifeLog.
Truthfully, I think it’s a bit of a show, as the Cabal doubles down on surveillance and takes another step toward a social credit system that will make Orwell’s vision look idyllic in comparison. Think Black Mirror season one, episode two. They’re piloting the program in China because a totalitarian regime can get away with anything.
I also think there’s a connection to the traitorous shuttling of technology, data, and weapons systems from the U.S. to China (i.e. Google, Facebook, Clintons, Feinstein), but I haven’t put all of that together.
I’m so conflicted on all this. :). I love remarketing!
Great Article by Justin Brooke on Medium About the 'Genius Pixel'
I mentioned this article earlier. I thought it would be a valuable addendum to the Retargeting Recipes review.
Justin premise is … strive to build “genius pixels.”
“A genius pixel is an ad account pixel that is so well trained to find your perfect customers or leads, that you don’t even need to use keyword or interest targets anymore.” --- Justin Brooke
Back in the day advertisers had to prospect with brute force, bringing in a ton of traffic to an offer, hoping a decent percentage would convert. Brooke says that’s changed now with the exponential improvement of AI and machine learning.
“Would you like to create a genius pixel that hunts down just your customers, lowers your cost per sale, and increases your ROI?” --- Justin Brooke
He drops some wild suggestions in that article, including ...
- Create a separate account for each product, particularly if they target a distinct market segment.
- Create a mini site for each product.
- “No single ad should show for more than 72 hours” (caveat: unless the ad keeps killing it over and over and over for weeks and months).
- It’s mandatory that ads start and stop for freshness. Campaigns and ad sets should not stop.
- 100 ad sets spending $5/day are better than one spending $500/day.
- “Facebook’s algo optimizes better with lower budgets.”
I agree that Facebook optimizes better with lower budgets, but once it locks in, you can gradually raise the daily budget without losing algo effectiveness. I’m sure there’s a cap on that spend because there’s a finite number of people in the market, but you don't need to build 100 ad sets, when one will do.
Keep the Ads Fresh and Your Sanity Intact
Regarding the “fresh” ads, Brooke does suggest using tools, such as Qwaya or Adespresso, to automate mixing and matching different components. Also, he says RevealBot is a great tool for automatically turning off underperforming ads.
These tools cut down on the grueling manual labor of cranking out so many ads and meticulously tracking the effectiveness of different elements.
Creating so many accounts, campaigns, and ad sets makes overlap a real concern. You don’t want to compete against yourself in the auction. In fact, I think Facebook and Google are getting so smart that you can give them the keys and they’re going to find the leads or buyers without having to do the extra work.
For example, I’m using Campaign Budget Optimization in Facebook. It’s working fantastically. Instead of competing ad sets, Facebook AI is taking care of it all, so there is no overlap, and I’m getting the cheapest lead cost possible.
And, Google Target CPA is similarly effective. Just tell them what your preferred result is and how much you want to pay for it on average and the machine will go out and find those people.
I’m on a Justin Brooke kick. I may not agree with everything he says, but he certainly makes me think. Consume as much of his stuff as you can. You’ll be a better digital marketer for it. Retargeting Recipes is a fantastic reference.
Check out my post about Justin's tests with the Pay What You Want (PWYW) model.
Recommended Stuff from Justin Brooke