Your digital advertising accounts used to perform a lot better, but these days you’re just not seeing the same returns. Should you cut back or hold the line? What factors are reducing your performance? Is there anything you can do to turn this around? In this session, we explore these questions and deliver actionable insights to help you improve your advertising returns, whether you manage those accounts yourself or work with an outside agency.
Presented by David “Psy” Deppner
Meet Magento Florida, February 1, 2024
TRANSCRIPT
All right, so this is the tech track. There will be some technical parts of the presentation, but we’re a little crossover here. I suppose. But that’s what the e-commerce world is. So we’re not in a recession right now officially. But there are a lot of businesses that are suffering. They might be you know, we’ve heard phrases like a, you know, recession in that just affects certain industries, right?
Sort of a rolling recession that’s not affecting the entire economy at once. We also hear a lot of talk about maybe we’re gonna have a soft landing from all the interest rate issues this year and we don’t know. Right? Maybe we will, there’s a lot of risks out there. So a hot topic with a lot of people right now is thinking about how to respond to economic downturn.
And there are a lot of advertisers right now, a lot of Magento merchants that are struggling, that are in the middle of a downturn that’s affecting their business.
So Brent introduced me already, but just a little bit of background, before running a paid ad management company, I was vice president of I.T. and e-commerce for almost a decade. We launched a Magento 1 site in 2008. We launched the Magento 2 site in 2016. So I’ve been on the tech side of things, been on the business side of things, just crossover between the worlds.
Psyberware is a small agency that specializes in ad management just for ecommerce merchants. Probably about half of our clients are Magento merchants. We deal with all aspects of ad management, but also with interesting integration issues between the ad platforms and the Magento site itself and the other e-commerce platforms we also use. So the agenda today is going to be pretty simple.
I’m going to talk about some things about facing reality. What does it mean to be in a downturn. Right. And what are some common responses and maybe what are some problems with those responses. And I want to talk about some optimization. And here’s where we’re going to get a little bit more into some technical weeds. Right. What can we do to make the Magento site and the ad platforms work better together.
So the ads perform better, and we squeeze a little bit more juice out of that lemon. At the end I want to talk about just diversifying a little bit, trying some new things, setting ourselves up for future success as we come out of whatever downturn we’re experiencing. So first off, let’s face the music. If a business is going into a downturn and they’re seeing a real slump, it’s going to be hard right?
When people start to see their business results slack off, they usually follow a couple patterns and sometimes it’s one after the other, right? So frequently the idea is, “Oh no, the world’s crashing. Our results aren’t as good. We need to preserve cash, we need to cut expenses.” And so there’s a knee jerk reaction to look at what the biggest expenses are and to cut.
And oftentimes advertising is one of the biggest expenses. But there’s also another reaction that is, “Let’s buckle down and really try hard and do better and get the results that we got last year. Let’s make sure we solve the problem that’s occurring right now and get our get our numbers back up. And if we don’t get the numbers back up, some heads are going to roll.”
And so frequently we do one of these. And then we do the other of these. And they all get combined into a big mess. So let’s talk about what happens when you cut ads entirely. You see that ad budget is a major expense. And you say, we got to ax that in order to save cash. There’s a real example from a real merchant who faced that at the end of last summer.
So early on in the year they were growing, their costs were going up. The revenue was also going up. If you notice, by comparing those two graphs, though, once we hit summer, revenue started falling off a bit relative to the costs. Costs kept climbing. That’s interesting. That business is definitely experiencing a little bit of a downturn in their industry.
They know they’re in a industrial downturn. Their competitors are also facing a similar issue. Their suppliers are saying everybody’s having problems. They made the decision to cut the ads entirely. And you can see ads cut revenue cuts. A few weeks later, they realized that was a mistake. They reversed course. They turned the ads back on. The reason this was a mistake is they were in a scenario where they were like basically spending a dollar and getting $3 back.
So they thought they were going to save, you know, a whole lot of those dollars they were spending. But then they lost a whole lot of those $3 they were spending in ad generated revenue after accounting for product costs and ad costs. And their cash flow got a lot worse. All of a sudden they didn’t save any money at all.
They made it worse. So then they started to try to ramp back up and note that the ramp up goes slowly, because we’re dealing with systems where there’s machine learning algorithms, machine learning models that have learned what works for your ad accounts. And when we stop the training data going in, they no longer have accurate current data and they have to reset and start over.
And you’ve also changed the dynamics with your competitors because you just gave them a huge gift. They just got a boost in their revenue. Their ads are performing better. So they’re spending more now and now outbidding them to get those clicks is that much harder. A way to think about profitability in the ad systems is that there’s this interesting response curve where as we spend more up to a point, we make more, and then past that point we make less.
There’s a point of maximum profitability. The point of maximum profitability really has to do with the crossover of when the increasing marginal costs of outbidding your competitors for more clicks crosses the point where you’re no longer making profit on each additional product you sell, right? So there is a point where you’re making the absolute maximum profit that you can on a given ad account on a given campaign.
And where this is, is different for different businesses, but there is a point where you’re maximally profitable. If you’re at that point on the graph here, B you’re doing great. If you’re up at A, you’re overspending on clicks that aren’t generating profit. If you’re down at point A, you’re underspending and leaving money on the table. So in a downturn, what happens is the opportunity available to you.
It shifts, it diminishes. The curve changes. And if you were spending at that point B before, you know that optimal point of profitability is no longer at the same budget level, you have to cut to stay optimally profitable. But you’ll be generating less profit than you were before. But any other point of spending, any other budget amount is actually going to cost you more.
So that’s the best choice to make if you’re optimizing for short term profitability. So think about the idea of maybe accepting the middle ground here. You don’t want to cut ad budgets entirely. You don’t want to turn advertising off to save money. But you’re probably going to have to cut your budgets right. You’re not going to achieve last year’s results because that opportunity is not available to you right now.
We’re in a competitive landscape. We have to outbid competitors for results. They’re all struggling with the same thing. You’re not going to get that. So instead, there is a gem of wisdom there about trying to optimize everything you can and squeeze everything you can in terms of performance out. And that that is a very good gem of wisdom.
Like, we want to do that, we want to optimize as much as we can. So let’s get the best results we can. Let’s think about some things to squeeze some more juice out of lemon. I’m going to talk specifically about a few issues that are common with Magento merchants right now, and are a little more on the technical side.
These are things that dev agencies could help their merchants with, ad agencies could help their merchants with, and merchants who are doing things in-house, could do a little bit better on. So number one, a lot of advertisers have far too many campaigns. And again, think about the fact that ad campaigns are really using machine learning models to figure out how to bid based on a whole lot of different factors.
Those models feed on data. We need to give them the most data possible for them to perform the best that they possibly can. When we fragment ad accounts into a whole lot of really little tiny campaigns, we’re splitting up the data in a lot of little buckets that the campaigns don’t optimize as quickly, they don’t optimize as well.
And there are frequently opportunities to optimize just by merging campaigns that are redundant or don’t need to be separated in together. There are reasons to split campaigns separately on accounts, but a lot of campaigns are split separately just because some manager in the business wants to have data on separate line items when they’re looking at a list of campaigns just for reporting purposes.
So I always think it’s important to think about driving performance with your campaign structures, not setting them up for reporting. There are other ways to report on the data.
One of the really good reasons to split campaigns out separately is when you’re dealing with a merchant that has products that have a whole lot of different gross profit margins. There are some merchants where all of their products are 40% margins, but most merchants have products that are 25% margins. And then we’ve got a cluster of products around 35 and we’ve got some that might be 50 or 60% or 80%, right.
We should be treating these products differently. And if we’re feeding these out in data feeds out of Magento, and we actually can flag them based on profit margin levels, we can segregate them into different campaigns that are bid more or less aggressively. And this is just another example of that idea that if we’re, you know, we want to not leave money on the table for the really profitable products.
We want to spend more to get more. But the really low margin products, we don’t want to spend as much. We don’t want to overspend and lose money on those. There’s a place where we can set our targets, where we’re maximally profitable. So here’s an example of a business that has budgets that are fairly flat throughout the year.
Okay. And then at the end of the year, you can see a spike up for holiday sales. They want to spend more on the holiday season. And the same principle applies to this sort of scenario. So the graph on the left here is budgets over the year. The graph on the right is performance over the year: ROAS, right?
Note that summer has an awful lot of performance. You know springtime into summer. That’s where there’s an opportunity to make a lot of money that’s not being met by raising the budgets. Later in the year, actually, even though we’re spending a whole lot more in the holiday season, because that’s when people are shopping, right? But that’s also when competitors are spending more.
So there’s not actually a proportionate lift in the results late in the year. So if you do an analysis on something like this and instead split the budgets so they’re spread differently throughout the year, you know, just spending the same annual budget but flex throughout the year in a different way can easily give you another 5% revenue for the same spending.
All right, tech issue here. Most Magento merchants are using a very simplistic feed generation tool. They’re using a very simple extension that came with some built in templates, and they’re just plugging the Magento product attributes into this template. They’re uploading that to all the different platforms that they’re working on. Right. There’s a lot of opportunity for improvement by implementing all the other things that you can do in a product data feed.
So most Magento merchants still haven’t implemented product attributes that Google added to their product data specification in 2020. They’re still using, you know, maybe they’re adding ten attributes into their data feeds. And if they put some work into it, they could really flesh out the data. They could be putting 20 or 25 attributes into their data feed per product.
And that would help with product matching to user search queries. Yeah, obviously more sales would come from that. Higher conversion rates. A current issue right now is third party cookie deprecation. We are losing the ability to set a cookie on a user’s browser and then add them into an audience and show them remarketing ads or, you know, bid higher for them when they search for our products in the future.
So one of the things that can be done now is to create audiences based on customer data, where we take some, some token, like an email address, do a 256 hash on that, and we submit a list of these hashed email addresses to Google or Meta or LinkedIn. And we create audiences of people who are past customers or our best customers or people that are on our email marketing lists that that we’re communicating with on a regular basis for email.
So setting up audiences like this can give us the ability to communicate with our customers as we lose the ability to to put them in audiences with cookies. And that’s rolling out right now. Right. So this is a current issue that will have a huge impact on performance this year. Enhanced conversion tracking uses a similar technology to hash the user’s email address or some other information like that, and pass that along with the conversion data.
So that hash data is a enables better cross-platform conversion tracking. It also enables conversion tracking in situations where some of the data is being blocked or even when they’re, you know, blocking conversion tracking with browser extensions, enhanced conversion tracking can lead to a few percentage points more, conversions being tracked. And that might not sound like much, but that might be a little bit more later in the year due to cookie blocking issues.
Also, if you’re just a few percentage points more efficient in feeding data into those machine learning models, they get better than your competitors. And there’s a snowball effect over time. Over on the Meta platform, it’s even easier. You don’t even have to set much up. Yeah, I mean, you can just flick a flip a couple of checkboxes and the Facebook Pixel will pay attention as people are going through checkout and they realize the person’s typing in an email address.
I’m just going to grab that. It’s going to be hashed in JavaScript in the browser, so it’s never shared offsite with anybody else. But then that hash value passes over to Meta, and they can attribute conversions in situations where they would otherwise be blocked and be more efficient at running the campaigns in the future. Server to server tracking is something that almost nobody has implemented across all their ad platforms.
Some people have set up the Facebook Conversion API. It’s a really interesting thing where we’re still doing the front end JavaScript conversion tracking, but we also have a server to server channel on the back end. It doesn’t go through the browser at all. It can’t be blocked. We pass some of that same information, some hashed tokens about who the user was, some information about what was in the shopping cart, the value of the purchases, when we can maybe pass tax, shipping, other information like that, and that data on the back end on the ad platforms can then be patched in to improve the training of the machine learning models.
Okay. So just real quick, some other low hanging fruit. How about product disapprovals? Do any of you have like 10% of your products in merchants that are disapproved? What about the Facebook catalog? I mean, fix the data issues there and you might have a 10% boost in revenue overnight. Brand advertising. There’s a lot of companies that think that they shouldn’t advertise on their brand keywords, because those people are going to come buy from them anyway.
Go check the click through rates on people that are searching for your brand keywords, and see how many of them aren’t actually clicking on your organic result, right? There’s that. It’s an issue. Your competitors are advertising on your keywords. And maybe you should too. So how about SEO? SEO is a good one.
So officially, Google says your organic search results will have no impact on your advertising. That’s true. They don’t directly have any interaction. But there’s this sort of side channel that’s interesting. If you have organic results and paid ad results on the same search engine result page, your a user is more likely to click on those.
You’ve knocked an extra competitor off that search result page. You have higher click through rates. Click through rate is one of the key factors that drives the quality score mechanism that Google uses to determine how much you have to pay per click to win the auction. So if you get a slight quality score advantage through the increase, click through rate, you get cheaper cost per click.
That’s a financial factor that’s going to influence your bottom line and, influence the ROAS of your ads. So we’re here at Meet Magento, Florida, right. There’s a lot of opportunities to explore other website improvements. The thing to think about is every website improve ment that improves the website’s conversion rate is going to make the ads more profitable automatically, without spending any more money at all.
Right. It’s a way to squeeze a little bit more out of it. So going to talk about just a few ideas about how to diversify. And then we’ll wrap it up.
If we think about what to do here. What else can we do. What can we try to do that we’re not doing already? Other ad platforms, other opportunities. It’s hard because every business is different. Like our our, our customers are on different platforms. And,
There’s no common answers that are true for every business. So I’m just going to give a few thoughts, maybe talk around some of the issues. First off, maybe think about focusing on that lower funnel. If you’re in a downturn, this is the time to think about platforms where people are actually searching for products. Some vague brand building advertising on social media that might pay off in the future.
Maybe that could go for a while. We can turn that back on next year, right? A lot of merchants are advertising on Google. Almost everybody. But a lot of people don’t advertise on Microsoft. And that’s that’s a little strange to me. It’s really easy to set up a Microsoft ads account. People on Microsoft are searching for products.
It can be really, really profitable in some industries. Like a lot of people have windows computers that are locked down and they can’t upgrade their search browser, right? They’re the browser, so they can’t change their search, all that sort of stuff. So they just default into using a Microsoft product and ending up on Bing search. A lot of government workers, a lot of people in industrial situations where they can’t change anything on their computer, can be really useful in B2B contexts to advertise on Microsoft ads.
A lot of Magento merchants also don’t sell on Amazon, or if they do, they have a resistance to advertising because they’re already paying so much. And all these extra fees to Amazon. But you can set up some really simple Amazon campaigns that are bid at a really low level and target them for a really high return on ad spend and increase your Amazon sales fairly easily.
The fact that a lot of people don’t want to do that actually is wonderful for those of us who do. There’s a lot of opportunity there. And when you think about what to try, think about, well, you know, if something’s working well on one platform, is there a way we can duplicate this to other similar platforms?
So if you have search data on Amazon ads about what people are looking for there and converting on those search terms might work well on Google as well. Right? If you’ve got some things that are really hot products that are working somewhere, maybe, you know, on your Google account and you know, what’s selling really great and what’s the hot thing?
Well, setting up some social media ads that are more direct response and not the fluffy brand building awareness stuff that’s going to get you results down the road. But this is the product that’s selling right now, today. Get some direct response social media ads going on those key products, you know, set up some carousels with your top sellers.
There’s always opportunities like that to expand what works on one platform to others. I would say when you’re figuring out what to cut and what to expand on, just be really careful about social media attribution models and how they count their conversion tracking and what’s going on with them displaying products on audience networks. We frequently have situations where where ads are shown out in, you know, on third party websites, when you’re running ads on another platform.
And they they’re attributing conversions based on people just passively viewing that ad and not interacting with it at all. It’s a huge cause for double counting conversions, so it’s usually good to be skeptical about those numbers and really understand what they mean. So what else can you try? Social media is really great for narrowing in on particular audiences.
There’s obviously a huge shift the last few years to short form video. If you’re in a downturn and you don’t have a lot of money, but you do have time, you might want to just play around with that. You might want to spend some more time on platforms that you don’t normally use just as a user, and pay attention to the ads on there.
What resonates with you? What looks like it’s working? Think about where your customers are moving and stay in front of them. So we’ll wrap it up. Remember what I said about those two most common responses to a downturn earlier on? It’s like either we’re going to slash the ad budgets to save on expenses, or we’re going to try to achieve last year’s results by trying really hard and firing people that can’t pull their weight because they’re not performing anymore, or, you know, some combination of all of that.
Your competitors are going to make these mistakes, and that is actually the best possible outcome. If you avoid these mistakes and your competitors make them, then you’re in that wonderful situation. Your competitor dropped their ads to nothing, and you get this beautiful spike in sales. It’s lovely. Happens all the time. Or they fire the key people in their marketing department that couldn’t perform.
And now, coming back out of the recession or downturn, their best people are gone. Those are the people that got them their growth during the last business cycle. Think about that. There’s a huge opportunity. But the biggest opportunity is, as you’re coming out of that. So to wrap it up, what matters the most? Face the reality that you’re probably going to have to cut.
You’re probably going to have to spend less and you’re going to make less. But don’t turn the ads off right. Optimize more for short term profit. Optimize for lower funnel ad results. Modernize all those ad integrations like a lot of those things aren’t going to make a night and day difference in your business, but an extra 3% performance here and 6% performance there and 5% performance there on different things that you do that adds up to real money when you make a lot of those improvements.
Really look into server to server conversion tracking as there’s a lot more ad blocking on the front end. This is going to be increasingly important right now. It’s kind of hard, but if you can get ahead of the curve on that, you’re going to be ahead of your competitors and seriously experiment and try some new things, especially if you are in a downturn and you actually maybe have some time and you don’t have as much money.
Try to invest some of that time in some experimentation with low dollar experiments and see what you can do. This is all about getting ready for the rebound. So as economic conditions improve, you’re able to really grow as the market moves back up. All right. That’s all I got for you today.