#142: Facebook Ad Troubles: Understanding The Drop And How To Fix It
This episode of the Ecommerce Coffee Break Podcast features a conversation with Brendan Hughes, CEO at Optily. We discuss why Facebook Ads performance has fallen and what can be done about it.
On the Show Today You’ll Learn:
- How did Apple iOS 14.5 affect Facebook ads
- Is Facebook advertising still worth it
- How to shift budgets in your ad campaigns
- What is an ads trading desk
- What is Optily's role in helping Shopify merchants with paid advertising
Links & Resources
Shopify App Store: https://apps.shopify.com/optily
About Our Podcast Guests: Brendan Hughes
Brendan Hughes has worked in eCommerce marketing for decades on both the ad-selling and ad-buying sides of the business. Throughout the years, he’s seen the issue of walled gardens and siloization of data from platforms such as Facebook and Google come up as a problem again and again. Since 2019, Brendan has led the team at Optily where he is focused on solving this problem for Shopify merchants. Optily self-service app enables merchants to optimize their cross-channel ad spend instantly across campaigns to grow sales.
Listen & Subscribe on your Favorite Podcast App:
Apple Podcasts | Spotify | Amazon/Audible | Stitcher | Google Podcast | YouTube
Please support the show if you liked today's episode:
- If you love the podcast, please get someone else to listen, too!
- If you enjoyed this episode of the Ecommerce Coffee Break podcast, please head over to Apple Podcasts, leave a rating, write a review, and subscribe.
- Share the podcast with your family, friends, and co-workers.
- Tag the podcast on Instagram @clauslauter and let me know what you like about it.
- If you like the content and would like to support the podcast, you can buy me a coffee here.
- Become a guest on the show or sponsor an episode.
Claus Lauter: Hello, and welcome to another episode of the E-Commerce Coffee Break. Most of online merchants using paid ads, most of them use Facebook ads, Google ads, et cetera. Now, the last year has been a bit of a blog pass for a lot of them, and today we want to find out and discuss why Facebook ads perform and has fallen so much and what you can do about it
Brendan Hughes has worked in eCommerce marketing for decades on both the ad-selling and ad-buying sides of the business. Throughout the years, he’s seen the issue of walled gardens and siloization of data from platforms such as Facebook and Google come up as a problem again and again.
Since 2019, Brendan has led the team at Optily where he is focused on solving this problem for Shopify merchants. Optily self-service app enables merchants to optimize their cross-channel ad spend instantly across campaigns to grow sales.
Brendan Hughes: Hey, Claus, thanks for having me on your podcast. Welcome,
Claus Lauter: Brendan. Gimme a bit of an update. What last year for I 14 five for the [00:02:00] ones who do not really understand what kind of impact that had coming afterwards.
Brendan Hughes: Back in April, 2021. Implemented some privacy changes, some consent changes in for iOS users and primarily obviously on iPhones, and really asking people to opt in for sharing data , with certain apps and with all apps.
And in particular, this was targeted at, Facebook and Tim Cook at the time kind of made very public statements about how he, , dislike. Facebook's approach to data management and privacy. , and we also suspect that Apple has its own long-term game around advertising and kind of generating its own revenue.
So there's always a kind of a few sides to this and what that meant. Was that, , a few things happened at the time. , so one, Facebook changed some things, , from, particularly in terms of how it gathered data and how it managed data, , in order to soften the kind of the negative perception around itself and to soften, , to be more amenable to some of the privacy concerns that, that people had.
, so we'll kind of [00:03:00] recall that, at the time, , it reduced the number of events. , that you could track on your website. So there was, , that was reduced down to eight events. And previously if there was a limit, it was a very high limit in terms of , the number of events and people never ran into issues.
So it meant that there was far fewer things you could track from kind of users. And these weren't just for iOS users, , or Apple users. These were for all users. So they made these changes globally in order to soften the impact. Then they also looked at the attribution windows.
And so they stopped storing data for as long a reporting back on data. , so previously you were able to kind of have longer attribution windows alone. So look back windows in terms of when somebody first saw an ad or interacted with an ad and then went on to purchase. So there were a number of changes that.
We're better from a privacy perspective. We're more protective of, people's personal data. But actually that's negatively impacted on the performance of ads. , so what we saw instantly was. Confusion from people who were advertising on, Facebook and Instagram in terms of they needed to do [00:04:00] some things.
They needed to verify their websites, they needed to change their pixel settings. , and suddenly they had, , this dramatic drop in performance. , and at the time, initially people didn't really understand. Was this a real drop or was it just a data loss? , but ultimately we work in the data game. , we all depend on data and over time as actual store sales in particular, maybe in the Shopify world, as those store sales maybe declined.
, then people kind of understood that actually the performance of these channels was actually declining. We were able to see both directly attributable sales. Those numbers had gone down, but then overall sales started to go down. So we started to lose confide. little bit in, these channels.
I think the other thing that then happened, cuz if you think about it, we were coming out of the pandemic, , a lot of brands had exited from advertising during the pandemic. So in particular travel brands and, you know, hospitality and tourism brands , and e-commerce had boomed, right?
So ad spend in the ad channels had increased. Due to e-commerce in particular. but then all [00:05:00] the travel brands kind of came back on. All these other brands that had exit out of, kind of advertising came back in. , and what we started to see was the prices just going up so.
In Q3 2022, you're probably now paying three times more cost per impression , in Instagram for a brand awareness campaign anywhere two years ago. Essentially supply and demand kicked in. The supply of users didn't, particularly in western markets. So Facebook growth has flatlined.
I. Growth has, you know, has continued to grow, but not so much , in western markets. So probably more , in eastern markets and maybe developing markets. Availability of, , users and audiences in these platforms didn't necessarily grow probably as quickly as the demand for audiences has grown amongst, advertisers.
So there's a few things that have happened. together for Facebook and Instagram and it's still a great platform, right? It's still a great place to be advertising and finding audiences. , it's just got a little bit harder than it was maybe two years ago.
Claus Lauter: Okay. No, I think that was a very, very great [00:06:00] summary there.
And I get often asked as like, should I still spend money on Facebook? And my answers.
So don't put all your money horse. Not only Facebook go also on other platforms. Obviously there has Google, it's around for 20, 30 years. , TikTok is coming and so on. But now it becomes really difficult for a lot of marketers to find out to attribute.
Brendan Hughes: The difficulty here is that, The pixel, the attribution that what we were relying on from the Facebook pixel has got weaker, there's less data kind of now available to us. We probably never fully trusted the data that comes from ad channels anyway.
And it's not just Facebook, but it'll be other channels as well. And really everybody's been struggling with the attribution topic for a long time. , and there are other sources of data, so mostly we're using maybe Google Analytics or maybe we're using the analytics, , and conversion kind of attribution that's inside a Shopify in particular.
And then there are other great tools. , that are helping with attribution. Very often they're what we would call pixel based or cookie based. So they are putting their own pixel or cookie onto your [00:07:00] website. , and the challenge there is that. A lot of what happens inside of the platform. So we think of Facebook and Instagram.
A lot of the user behaviors that happen, , are not trackable by third party pixels. So if, , somebody sees an ad but doesn't click on it, , and then maybe a day later or on a different device comes in , and buys that product because , maybe they've seen another ad or a remarketing ad or something.
, then of course no third party pixel tool , can really kind of spot that. , so one of the approaches , that certainly we've adopted and we're seeing people adopting is, not relying on any one data source. So leveraging, , the data that is available inside of Facebook and Instagram. ,
for what it has. And we can pretty much rely. On that behaviors inside the platforms, , is very reliable , inside of meta. If somebody's seeing your ad, , somebody's clicking on your ad, somebody's sharing an ad or liking it, et cetera, all of those data points are very reliable inside of the platform.
It's when somebody leaves the platform and [00:08:00] then goes , to your website or to your app, to do some research or to make a purchase. What we see people doing is making sure that , the technical integration between, , let's call it meta, for example, your Shopify store, that that's been done correctly.
You have verified your store between Facebook and Shopify. You have implemented your pixel and your pixel tracking is in place, but you've also implemented, , what Facebook calls it's conversion api, , or it's server to server tracking. So bypasses the cookie blockers at the pixels.
And it means that when somebody makes a purchase on your store, then your server is sending that conversion information back to Facebook and. Meta does is it combines those pieces of data. So if it has pixel data, it'll use it. If it has conversion API data, then it will use that and it'll, , duplicate as well.
So it's not double counting somebody who fired a pixel. And then also is recorded on the conversion api. I, and then the other, reliable piece of data that , , we see is the, Shopify [00:09:00] conversion. And why that's interesting is that it's based on UTMs. , so UTM parameter is a, string that you append to the end of your URL when you're in them.
, you're setting up an ad campaign or an email marketing campaign, so it is going to be click based. It does rely on the, on somebody clicking through. But Shopify is essentially using what we call a first party cookie. It's not blocked by kind of cookie blockers and Shopify is then tracking that user, , on your store.
if we chat a little bit about attribution, kind of windows and attribution models. Okay. So at the moment, Facebook only reports on and only technically records if somebody. Seen an ad and clicked on that ad in the last seven days, , whereas the Shopify attribution window was 30 days. So if somebody clicked on an ad in the last 30 days, , and then has gone on to make a purchase within that kind of timeframe, then you'll see that data.
So neither [00:10:00] is perfect cuz one is Shopify's based on clicks, , only, , and it's looking at first click and the first thing that somebody does, and the last thing that somebody does Facebook is based on impressions and what and what we call view through. So somebody kind of sees something and then separately goes and, buys or interacts with your store.
, and it's only got seven days of data. But by actually putting these data points. And starting to, , , let's call it, , either correlations or patterns, right? So, , something's happening in Facebook. You're spending some money. , you're seeing trackable, directly attributable, , sales and, and actions on your website, but then you're also seeing the overall picture increasing.
, and there are some new and interesting ways that we can measure that as well, that make life kind of easier.[00:11:00]
Claus Lauter: Okay. Now, a lot of small, medium enterprises, people who are not IT related , in their business, they already can feel it's a very technical topic and so on, so forth.
Now, one thing that smaller enterprises, brick and mortar stores going into e-commerce, , don't have as time or resources to become a data visit and. I understand that LY helps with that, that you have developed a tool to make their life easier. Tell me a
Brendan Hughes: little about it. There's multiple sources of, data that we can look at.
What we aim to do just with our, Shopify app is [00:12:00] to join those kind of data sources. So when you install our app, , and it's, click, click several, you know, two, three clicks to install, , you connect your ad accounts, , inside our app. What we do is we line up what the ad channel says is about a campaign.
So Facebook says, Hey, , we delivered a ROAS of, you know, five. We then look at what Shopify attributes to those campaigns or to Facebook in total. And we're not trying to say one is better data point than the other, but what we're looking at is , when those data points change over time, , then that's interesting.
So the Shopify ROAS is declining. To look at here or, , the cost per click on Instagram is increasing. Okay. Something to look at there. There's two different types of measures. There's the short term measures, which is, , looking at the attribution.
Data points that are available from these different sources. But then there's also the long term, data points and we've seen a lot of research and a lot of evidence from, our clients , and across the industry that says, look, we need to be investing not just in short term results, but in these longer term results.[00:13:00]
And one key metric that, , we expose kind of automatically in our platform is, , marketing efficiency. This is where we're looking at the total store sales. So not just those directly attributable to advertising. And we're taking that as a ratio against ad spend. So for example, we know what a ROAS or return and ad spend is, , the total sales attributable to advertising as a ratio against, ad spend.
So if you have a robot of. Right. So that means that for every $1 of ad spend, you're able to identify $5 of revenue that has come from the ad channels If you have, for example, an A marketing efficiency ratio or an MEOR of 10, then what that means is that for every $1 of ad spend in total, your store is generating $10 of of revenue.
And why is that useful? Because we're kind of saying, well, hold on a sec. A lot of that $10 is not coming from ads might be coming from other places. So it does a couple of things for us. One is it helps us to look. Just a bigger [00:14:00] picture of, how much can I afford to spend on ads? Let's say my gross margin is 5%, then I can't afford to have a merit of 10 because I'm spending 10% of my gross revenue , on advertising.
So that's , one thing that helps us , to guide against and let's us understand actually our business is scalable or not scalable because of, where that meor is sitting. It does give some indication, especially if we look at trends over time, , of the directly attributable kind of sales versus the non directly attributable, because we do know, we all understand that, , some element of what we're doing in advertising has a longer term impact.
And there again, there's research out there that would suggest that a matter released some in research recently with. and talked about about 60% of the impact of advertising is long term, right? , and there are things you can do to enhance that, but you know, so that's potentially suggesting that 60% of the money we're spending today we can't see today are in the next seven days or the next 30 days, depending on our attribution window.
So that's important. , and maybe the final thing around, , the marketing [00:15:00] efficiency ratio is, , it helps us to understand. How good we are at the other parts of our business, right? So how good are we at, not just converting, but retaining customers? Because there's a lot of effort that we would put into, , remarketing , to our customer via email, via s m s and via advertising.
, and so for example, we'll see that subscription kind of businesses will have a much higher m eor, right? Because they have this kind of automated repurchase, recycle, , or cycle in place. , , it's a really good indicator of the overall health of our e-commerce operations kind of more generally.
, and because what we want to be doing is, you know, acquiring customers is expensive, so we want to make sure that if we're acquiring them, then we're able to kind of maximize the lifetime value of those customers kind of beyond that initial purchase. I think that's
Claus Lauter: a very, very good point that you made there.
First of all, having statistical significance, have data over a long run. I was never a big fan of ROAS because , it's just a small picture of a bigger picture , and people tend to overspend [00:16:00] to scale too fast. , and basically at some point or very quickly, it backfires. With the m i r, , is there anything that also can tell you on how you can shift budgets then to different channels, say between meter and Google?
Brendan Hughes: That's something we, focus on. Okay. , so our application will look at , these numbers, , and what we allow you to do is to identify your different business objectives. So of course, everybody has primarily a sales object. If we think of , the marketing funnel and the marketing funnel, even though it's been around for, you know, as a concept for 120 years, it's probably even more relevant today because we can actually start to measure things kind of better today.
What we find is that, most of us start with our e-commerce marketing, we're absolutely focused on sales types, campaigns, catalog campaigns, , really trying to measure kind of CPAs , and roas kind of at what we call bottom of funnel. The problem with that is that, , it struggles to.
Right, because , where you get massive efficiency [00:17:00] on your CPAs, your cost per transaction or cost per acquisition and your robots is, , you're targeting people or addressing audiences who are. In market for your product have high purchase intent and are probably actively searching, which is why, Google product ads, , shopping ads will work so well.
, or Facebook catalog. Remarketing ads will work so well because you're already talking to people who maybe are familiar with your brand, , and are kind of at the point of making a purchase. What we often see people trying to do is. Take those types of campaigns and go out to broad audiences, right? So now we're trying to move up the funnel, right?
it just doesn't work because you're taking the most expensive campaign type. So if your C P A, for example, is let's say $15 on a average order value of 75, so you're getting a nice five x roas, right? And you take that type of campaign and you go to a broad audience, you're now gonna spend $15 in principle.
To people who've never heard of your brand, have demonstrated no purchase intent towards your product or your brand. What we, , are seeing a lot of more people [00:18:00] doing is then leveraging maybe traffic campaigns or awareness campaigns. , and these are all available in, the platforms, , going out to broader audiences.
They're much more cost effective. So for example, , if you imagine your c p a, as I say, $15, , if you are using a, for example, a video views campaign in Instagram or in Facebook, then you're probably paying, , for a 75%. Video view of a 15 second video. So somebody's watching maybe around kind of 12 seconds, which is a good indicator of quality given that most people exit after the first one or two seconds.
, you're probably paying kind of one and a half to two and a half cent per video view, right? So now instead of paying $15, you're actually paying a thousand times less, right? To get a human to engage with your kind of 15 second video. So you can now go kind of much broader. Of course you're probably using different placements in the platform, so you're using, you're not in the shopping carousel, right?
You're not probably highlighting your, product so much, but you're probably doing more brand storytelling and product storytelling. That's what we're seeing is that people are using different creatives. [00:19:00] Different optimization strategies and different placements, so different channels, for each of these.
, sorry, a long answer to your question. Clouds, right? But in terms of what we see people then, now what we facilitate people to do is to say, look, what's your business objectives of sales traffic visibility, right? , which campaigns are delivering for those objectives. So we organize things together across the different channel.
And we start to kind of line them up and kind of measure them against each other. So we'll have maybe , an Instagram campaign kind of lined up against , a YouTube campaign, maybe lined up against a TikTok campaign for top of funnel kind of brand awareness. And we're starting to kind of, I say benchmark them against each other, look at the performance over time, look at engagement metrics and say, Audiences from these campaigns we're able to get outta them much more cost effectively, but they don't engage as much or they don't convert ultimately down the funnel.
, and what we do is we surface opportunities in our application to start adjusting some of the campaign parameters, but also maybe to start [00:20:00] optimizing your budget. And moving them between kind of different channels and different placements and different different campaigns what you get from that.
And I look, I think as the prices are increasing in the ad channels, , what we set out to do is to really try and help you achieve those marginal kind of improvements from the campaigns. It's never a good idea to make dramatic changes in particularly around budget, on campaigns, on a day-to-day basis.
But it's really spotting opportunities. A bit like a trading desk, , but made simple, right? Because, , most of us are very busy running our, stores and trying to build out audiences, strategies and creative strategies. So the topic of managing spend. Can be hard. Right? , we kind of made a very intuitive interface.
We hope so that people can, , easily identify, , this campaign is starting to do very well. , there is an opportunity to spend more here. Great. , I'll make some kind of small changes to that today and then see how it progresses or This campaign is starting to dis.
so I'm probably wasting money here now on this because it's not performing as well as it did. , so maybe I'll reduce [00:21:00] the budget slightly, see if that helps a little bit. But also might need to look at my creative, I mean, might need to refresh it or maybe my audiences are too small, so I need to kind of broaden my audience.
So we also highlight those kinds of things for you. , but budget, And managing your budget can have a very dramatic immediate impact in terms of the performance of your and the ROI from your campaign.
Claus Lauter: I like the idea that you, , called it a trading desk. I was looking for a word like dashboard or something like that, but I think Trading Desk really sums it up.
So once people have installed your app from coming from opportunity.com, what is kind of the learning curve and how much time should they spend each day in the app? , to find out what
Brendan Hughes: works best? The first thing, and we've tried to keep it kind of relatively simple to onboard, but the first thing that people need to do is really identify their business goals.
We'll all have a sales goal. , and we see people initially kind of trying to organize everything as with a sales objective. , but I think when people look across the campaigns that they have live on the various ad platforms, they know that, quite a few of those are probably not,[00:22:00] Really measurable against kind of sales, you're gonna be measuring them against other things.
First thing that people need to do is just really organize, think about , their objectives and their goals and start organizing their campaigns based , on those goals. And then really we do kind of most of the work on a daily basis though our algorithms surface opportunities.
and those opportunities, , are really kind of identifying trends and patterns in campaigns, , that have been happening over. we allow, , new campaigns to settle. , we're looking over a seven day period, , and typically looking over a 14 day period to see what's happening.
So it's quite responsive, , because obviously , there can be a lot of changes in businesses what we suggest that people do is come in maybe once a day or once every couple of days and just review those opportunities and take actions and , some of them will be actionable directly within their Shopify dashboard.
They'll be able to kind of make some budget changes and kind of tweak those changes. Then others, they might have to go and think about their creative strategy. Maybe I need to refresh my creative and I need to get my designer, et cetera, involved in some but really what we're trying to do is reduce , that time, , from [00:23:00] the one to two hours of maybe analysis that people are doing, looking at different ad platforms on a daily basis and reducing that down to kind of the 10, 15 minutes of reviewing my opportunities and then making some quick changes.
Claus Lauter: Is a, , eight actual fire store. , tell me a little bit about the pricing structure.
Brendan Hughes: so philosophically, , I don't like to link an optimization or an efficiency solution with how much you spend. , so we've, , taken an approach that we charge you essentially based on how many ad platforms you've connected.
So our starting price is $49. , for example, you connect meta, which includes obviously Facebook and Instagram. If you add. Google, then it's another kind of $49 a month. , so each ad channel kind of is. , and that's our, introductory price is $49 a month per channel. And then actually after your first three months, then our standard pricing is $99 per channel.
So then it's $99 per meta. , and 1 99, if you add two channels and 2 99, if you add three channel. . And again, our goal there is that, , we're conscious that, , there'll be different sizes of customers who use their application. There are no [00:24:00] limits in terms of the amount of ad spend or kind of the volume of kind of transactions, et cetera.
The way we want to grow with our clients is, , as you said earlier on, need to increasingly diversify their channels. So we're very conscious that. , yes, you might start with one and, Facebook and Instagram are a great place to start because it can manage the full funnel for you.
, but as you grow your business, and if you want to be available and find your audiences, then you need to be on all these other platforms. And one of the most difficult parts about all of that is trying to manage your spend and your investment across those. I think
Claus Lauter: it's a very affordable pricing scheme that you have there, because keeping in mind if you go to an agency, there might only be focused on one platform, so there might be a meter agency, they will charge you a percentage of you spend and then you just see getting the numbers up and the end of the day, the quality, what you get out of it, the data might not be as good as with the app. , where can people find out more about
Brendan Hughes: Opti? If you go to the Shopify app store and search for Opti, o p t i l y, , or you go to [00:25:00] aptly.com, then you can find out about us, , there as.
Claus Lauter: Brendan, I think was a ton of very good information there for everyone and also to rethink their ad structure or their ad strategy there. And I'm a hundred percent sure even if the ad performance has formed, there's still a huge market and huge potential for ads out there.
Thanks so much for your
Brendan Hughes: time. Thanks for having me close. Thank you.
Today’s episode is made possible by our friends at BundleBuilder.
With 5 years+ in the app store, Bundle Builder is the OG bundling app for Shopify stores!
Bundle Builder creates custom bundles that will help you dramatically increase your AOV. The powerful discount engine builds offers and deals like %, fixed, or tiered discounts, plus you can easily get creative with custom product kits and gift boxes.
Let your customers choose their bundle contents for a personalized experience or set up pre-built bundles, one click and it’s in the cart.
Install today and schedule a call with our onboarding specialist. We look forward to bundling with you! More info and app download here.
Get notifications when new episodes are released. Unsubscribe anytime.
Join 3,000+ Ecommerce Marketers and Merchants getting our free Marketing Optimization Ideas to scale a growth-focused business on Shopify.
Continue the Conversation
In our ECOM MERCHANT PRO community, you can connect with our podcast guests and continue the conversation.
Our community is also a great place to get advice from other Shopify merchants who have achieved what you are aiming for.
This is your safe place to actively grow your online retail business with the support of the most amazing and helpful group of ecommerce entrepreneurs behind you.
Join the only Marketing Optimization Community where Shopify merchants go pro – together.
Running a Shopify business is tough, don't go at it alone.
DISCLAIMER: The content in this podcast episode is provided for general informational purposes only. By listening to our episode, you understand that no information contained in this episode should be construed as advice from Claus Lauter, idube Pte Ltd, or the individual author, hosts, or guests, nor is it intended to be a substitute for research on any subject matter. Reference to any specific product or entity does not constitute an endorsement or recommendation by Claus Lauter, idube Pte Ltd. The views expressed by guests are their own and their appearance on the program does not imply an endorsement of them or any entity they represent. The views expressed by the Claus Lauter, idube Pte Ltd, do not represent the views of their employers or the entity they represent. Claus Lauter, idube Pte Ltd, expressly disclaims any and all liability or responsibility for any direct, indirect, incidental, special, consequential, or other damages arising out of any individual’s use of, reference to, or inability to use this podcast or the information we presented in this podcast. We feature sponsored interviews with guests who paid an appearance fee that supports the maintenance of the blog or podcast. Affiliate links – if you click on my affiliate links and sign up for the products and services I trust and recommend, then I will earn a commission. Although we may receive a commission from the affiliate, the cost of the product for you will always be the same or often discounted. All affiliate products are vetted by me and my team, and we support and recommend these products because we find they are worth it.