Winning On Shopify With Retargeting
How We Finally Got Retargeting To Make A Profit...
I'm an eager student, but a slow learner. So when retargeting came along a few years ago, I jumped at the chance to learn how to do it. In this blog post, I explain Retargeting as a marketing strategy and how we failed repeatedly to make it work - and then ultimately found our way to a successful retargeting strategy. I'm not claiming to be an expert on this topic in any way - I'm just a student of the technique sharing my story...
Retargeting Defined: (in case you're not familiar with the phrase), "retargeting", it is the idea of placing an advertisement in front of someone who has previously visited your website. You do it by installing a "pixel" tracking code on your Shopify (or other) website that tracks visitors - so that you can serve up ads to them later.
(Note: Google calls this "remarketing", Facebook calls it "retargeting")
Because retargeting is an advertising strategy - the folks who offer you the "pixel tracking code" are generally Google (for their Adwords platform) and Facebook (for their advertising program). I'll describe a 3rd provider of pixels in a moment.
The Theory Of Retargeting: The theory behind retargeting is pretty simple. If you can serve ads to someone that has previously been on your website, chances are, they're more likely to click through and return to your site than someone who has never heard of (or visited) your site before.
As I've come to learn, this theory is something you have to work to prove correct (or incorrect) for your situation. Your testing can result in several possible outcomes:
- Theory Proved Correct: People who have been on your site before may truly be more receptive to advertising and return to your site (and buy) because of it. In this case - you still don't necessarily have a profitable advertising outcome. More on that later.
- Theory Neutral: People may actually not be any more interested in your site (or buying something) because they've previously been to it. In this case, it doesn't benefit you to use retargeting versus other traditional targeting methods (like age, gender, or interest).
- Theory Proved Wrong: In the worst scenario - people may actually be less likely to click on an advertisement from you because they have previously been on your site. In this case, the Theory of Retargeting is proved wrong - and for you- retargeting is a bad idea. You're better off advertising to brand new prospects.
Profitable Advertising: Direct response marketers are people that live and die by numbers. They do not speculate what will work for very long before they want to test it with real money, real buyers, real (good or bad) results. They test their way into new advertising approaches.
So when it comes to retargeting you have to test a few different things. I refer to this list as the COAT:
- Channel: Where are you going to put the advertisement? On the Google Adwords platform (text ad or image ad)? On Facebook (Newsfeed, right sidebar, Network, or Instagram)? Or somewhere else (I won't even discuss options beyond Adwords and Fb).
- Offer: What are you going to put in front of them? Just a general site ad? A coupon offer? An ad for a specific product, etc. What if none of those work?
- Audience: Which people exactly are you retargeting? For example, are you retargeting everyone that visited your site in the last 180 days? Or just the 25-34-year-olds?
- Type: What type of advertisement are you going to use? Text Ad? Image Ad? Video? Will one work better than the others? How many are you willing to test?
Marcus Lemonis in the Profit talks about "People, Process, and Product" but when it comes to online advertising the slogan must be something more like, "Common Sense, Competence, and Creativity."
Common Sense: The only online marketers that stay in business (or employed) are ones that can slog through the testing process sufficiently to justify their use of hard earned company money. If you lose money for your boss - why would your boss allow you to continue? If you're good at BS, then you might survive in corporate America with bad results, but not as an entrepreneur. If you are the boss, how can you justify wasting money that could have gone into other projects or your bottom line?
Of course, to mitigate the bleeding, you can test small (use just a little bit of money) and only invest when you find a winning method. But how long do you wait to find that winning method? As the naive farmer said staring at the huge pile of horse poo,
"There's got to be a pony in here somewhere!"
It turns out, there doesn't have to be a pony. You can simply burn money like an arson - and end up with nothing to show for it except a smoldering crater where your money used to be and a five-alarm financial fire.
Of course, the siren song of direct response marketing is a positive return on investment. The enchanting idea: Can you spend $100 on advertising and make $200? Can you spend $100 and make $1,000? (A 10X return). Truthfully, any return is a good outcome. Especially if you are building a customer base.
If you can learn to do this via advertising in support of your own business, then you are extremely blessed. In some cases you might not just find a pony in the pile of poo, you might find a unicorn that does circus tricks. Fame and fortune are sure to follow.
Competence: One of the biggest challenges with some marketing tactics, is that you can't actually try them unless you have the technical competence to set it up properly and ensure the data is telling you the accurate story. That is certainly the case with retargeting.
For many boot-strapping entrepreneurs, this is the stopping point. They might have money to burn on tests, but not have the talent to light the match (technically set it up). That's a hard hurdle to overcome.
Bring On The Consultants, Services & Programs
To solve this problem there are ready, willing, and (sometimes) able consultants, gurus, service providers, and experts ready to help you set up retargeting for a fee. Here are a few of the problems with working with consultants:
Higher Hurdle To Clear: A do-it-yourself approach is cheap but harder because you have to learn it all yourself. The other option, hiring someone to help, makes the financial success hurdle harder to clear. Because, to make the math work, you have to earn enough back in your advertising to pay for their help on top of your own internal expectations.
You're Too Small: Lots of gurus and services that do this work will candidly NOT be interested in working with you if you have a very small budget. So you won't be working with the best of the best. You might get lucky and work with someone that is a willing partner that is competent - if you can find them.
They Promise The Moon: Then there is the problem of these people selling you hard by promising you the moon, and delivery (you guessed it) another pile of poo ... or a smoldering crater where your money used to be, (at this point in the article you get to pick your preferred metaphor).
They Don't Know Your Industry: Finally, there is the problem of these people knowing (you hope) retargeting, but (usually) nothing about YOUR industry. So they don't understand the buyers, the products, the seasonality, or sales cycle, or the keywords used. They cannot tell you which product to focus on, which demographic, or which ad type.
This can, if you let it, be one-way cash is completely wasted by consultants. You are (in essence) paying them to learn about your industry, but guess what, they don't care very much since they work with so many people, they probably won't even remember the hard-won lessons the expensive testing provides.
So, for all these reasons, buyer beware.
Back to our list of fundamentals, Common Sense, Competency, and Creativity...
Creativity: All the items in the C.O.A.T. list above make it necessary for you to either be extremely methodical and design and then work through various tests with precise record keeping, (with near robot like powers). This will take time, record keeping, and money - and a robot like mental power. Sadly, I don't have that.
Or, you have to try and be extremely creative, (almost an intuitive level of artistry), and continue trying various creative combinations until you get something to pop. Even then, you'll need to remember what you've tried, why it didn't work, and what new tests you can attempt based on those lessons. This can be a hard road and generally leads to failure. I'm more of this type of guy.
For all the reasons mentioned above, I've learned that although retargeting (and tons of other cool marketing tactics) can work for some people (some of the times) it never means it will necessarily work for you (or me) any of the time. As the infomercials say,
"individual results may vary"
So the only thing you can do is test it. And for the last several years, my tests have fallen into three big milestones. Let me describe them and of course, in the third milestone, I'll explain in more detail why we got retargeting to work, and what it might mean for you as you work through these ideas for your own site.
I'll use our C.O.A.T. methodology mentioned above to outline these attempts.
Attempt #1 (Failure #1)
For our first attempt, we tried retargeting using the Google Adwords platform (they call it remarketing).
Although I wasn't savvy enough to set it up, I got lucky. Because we spend a reasonable amount of money each month on the Google Adwords platform, I have a personally assigned account rep at Google. It's actually a very nice service.
If you want your own helpful Adwords rep, just spend a couple thousand dollars a month on Adwords, then stop. Someone will call you from Google and say, "Hi, I've just been assigned to be your account representative - can I help you get your ads turned back on?"
So with their help, we installed the tracking pixel, let it cook, and began testing various ads. Here is the full C.O.A.T.
Channel: Google Adwords
Offer: We tried various general site ads and product ads.
Audience: We tried all sorts from the entire retargeting list, to subsets based on various factors.
Type: We tried both display and text ads.
Theory Of Retargeting Result: During this phase of our retargeting work we could get people to revisit our site and purchase something at a greater rate than "cold" traffic. So in terms of the Theory Of Retargeting - I had proven it to be correct.
But I was still not able to make any money, (positive return on my investment) in the campaign. In other words - retargeting worked, but my advertising effort still cost more money to run than it returned to me.
Well, I didn't know. So, I hired a freelancer off of Upwork (as I recall) to do a little audit and tell me why it was failing. He was actually a great guy - and super smart - and after helping me get a few technical things fixed, his prognosis was negative.
He said I didn't have a high enough average order value from the returning customers to justify the expense of the ads. In other words, I could get people to return to my site, but what they purchased didn't cover the cost of the advertising to get them there.
After pondering it for months - and trying various options - all without any improvement in the basic problem, I closed this chapter in our retargeting experiments and declared it a failure. Money and time wasted, but lessons learned.
Attempt #2 (Failure #2)
About a year later, a service provider that specializes in retargeting got me in their sales funnel. Or maybe I clicked on their ad, I can't remember. I won't name them here, since I don't want to be uncool. It wasn't actually a bad experience, they were very professional and nice.
They promised a lot. And they provided a lot of service too. One aspect of their offer that was a bit unique was that they could install a pixel that worked on both Google Adwords and Facebook. Here is the C.O.A.T.
Channel: Google Adwords and Facebook
Offer: We tried various general site ads and product ads. Various price points.
Audience: We tried all sorts from the entire retargeting list, to subsets based on various factors. This time I focused on Facebook in particular and the segmenting options they offer.
Type: We tried both display and text ads. But for Facebook, of course, it was various image ads.
Theory Of Retargeting Result: During this phase of our retargeting work we could get people to revisit our site and purchase something but ... this is interesting ... we couldn't get the retargeting list(s) to perform any better than a cold list of people, that we chose via Google Adwords based on keywords.
In other words, I was better off targeting people unfamiliar with my site... than I was targeting people that had been to my site previously. That was a horrible outcome for my retargeting effort and my ego. Prior exposure to our website was more likely to lead to a negative outcome than no prior exposure to our website? What?!
After months of fiddling around, working with this company, and trying various options, we concluded my retargeting effort was a failure. Money and time wasted, but lesson learned.
Attempt #3 (And Our First Success)
Our latest retargeting test started when we rolled out a new product. In an effort to spread the word, we figured we'd try Facebook Advertising - specifically advertising a video. The test is working very nicely. I am making more money than I am spending, and we couldn't be happier. It's not a unicorn that does circus tricks, but at least we aren't burning money like an arson.
Because I wasn't sure about the technical implementation of the end-to-end tracking, my awesome co-founder at Winning On Shopify, Kyle Hamar, helped me get it set up properly, (umm, I mean, he set it up properly for us). We already had the Pixel installed on the site, but what we also needed was the "goal" tracking to ensure we could accurately track a successful sale and attribute it to the Facebook ad. This was above my pay grade, but Kyle is a pro, so he figured it out.
The test is working very nicely. I am making more money than I am spending, and we couldn't be happier. It's not a unicorn that does circus tricks, but it's working. And, we are refining it daily.
After years of failure - getting retargeting to show signs of life is incredibly exciting.
How did we do it? We advertised a 3-minute video on Facebook selling the new product to people who have previously visited our site. I realize that sounds like a simple idea, but you'd be surprised how many years it can take to come up with a simple idea that works.
Here is the C.O.A.T.
Channel: Facebook newsfeed ads, (not right-hand side ads, not Facebook extended network, not Instagram).
Offer: We are promoting a 3-minute video that explains the new product.
Audience: We started our testing of the video ad to include both a retargeting list and also a list of likely buyers based on industry-specific interests. The retargeting list worked. The other lists didn't.
The retargeting list included prior visitors to our site in the last 180 days (as it happens we had just under 100,000 of them).
We are now refining our audience to exclude the non-responsive ages ranges. In other words, after we saw the ad working for some age ranges, and not for others, we stopped spending money to put it in front of the wrong aged people.
Type: A Video ad describing a new product.
Theory Of Retargeting Result: We are proving the Theory of Retargeting Correct. But more importantly, we found a way to make the advertising effort pay for itself (and then some). Our exact return-on-investment is TBD, but I can say confidently, it's a very nice multiple.
The Theory Of Retargeting is a hard one to prove at the individual website level and is not for rookie online sellers. A few months ago, I wouldn't have encouraged you to try it for the obvious reason that I couldn't make it work. But I'm becoming a fan more and more each day.
The lesson is pretty simple. Keep testing new things around the area of the C.O.A.T. and as you fail, try to learn something and devise another test. Keep trying using different videos, different products, different approaches. The hard truth is, it may never work for you. But like Thomas Edison and the invention of the lightbulb, after a lot of testing, you're very likely to find a formula that does work for you.
If you can use common sense, competency, and creativity, you've got a shot.
Now, I personally hate doing step-by-step 'how-to' videos, which is why I never do them on this site. But fortunately a lot of other people like doing them - so there are some great ones out there. Here are a few resources along those lines: