Welcome back to another blog post of Let's Talk Social, where we break down the ever-evolving world of marketing for business owners and savvy marketers like you. Today we're diving into a game changer, or should I say, a series of game changers: artificial intelligence and its impact on conversion tracking and digital advertising.
For new ways to gather insights into the privacy challenges rocking the industry, we're about to uncover how AI is reshaping how we understand consumer behavior. So grab your data report and let's get into it. First, let's establish the basics. Conversion tracking is how marketers know whether their ads work or not. Are people clicking, buying, subscribing? Historically, platforms used cookies or pixel-based tracking to monitor this behavior. But that's where AI steps in. With increasing restrictions on third-party cookies and privacy regulations like GDPR and CCPA, traditional tracking methods are falling short.
Enter AI-driven methods like digital fingerprinting and machine learning models that analyze behavior without needing personal identifiers. AI is enabling smarter algorithms that predict what actions users might take next. It's not just tracking what's already happened, it's forecasting future customer behavior. This means marketing strategies can shift from reactive to proactive.
So for example, platforms like Google Ads are using AI to offer enhanced conversion models, which estimate cross-platform engagement, even when data is incomplete. But it's not all smooth sailing. AI-driven tracking comes with its share of challenges and controversies. For starters, let's talk about digital fingerprinting. While it's a clever way to identify users through device settings or browser habits, critics argue it skirts privacy laws.
Remember the uproar when Google flip-flopped on their stance about digital fingerprinting? It's a prime example of how tricky this space has become. Then there's the accuracy question. AI models can predict conversions, but what happens when their predictions are off? Businesses can end up wasting ad dollars chasing phantom conversions, and we can't ignore ethical implications either. Many AI systems rely on massive datasets, often scraped from the web without clear user consent. Oh, and don't forget about transparency. AI can be a bit of a black box. Marketers might see results, but the how often remains unclear. That makes it tough to trust, especially when budgets are on the line.
Alright, so let's bring this home with real-world examples. Let's say you're running a campaign on Instagram to sell eco-friendly yoga mats. AI helps you segment audiences based on engagement history, past purchases, and even predicted interests. Instead of just serving ads to everyone interested in yoga, the system learns who's most likely to buy based on the patterns you didn't know existed.
Another example? Think about YouTube's ad tracking advancements. With AI, they're offering advertiser insights into whether a video leads to a product search, even if the user didn't click the ad directly. It's all about connecting the dots, even when the path isn't linear. But here's the kicker. While AI is great at finding patterns, it doesn't always nail human nuance. Ever had a retargeting ad follow you around for weeks after you've already bought the product? Yep, AI can be persistent, but not always smart. Wasting ad dollars.
So what can you do today to stay ahead in this AI-driven world of conversion tracking? First, embrace first-party data. With cookies on their way out, focus on gathering data directly from your audience through newsletters, surveys, or loyalty programs to help get around AI's miss on their tracking. Second, learn about AI tools. Explore platforms like Google Enhanced Conversions or Facebook's AI-driven optimizations. The more you understand these tools, the better you can leverage them.
Third, stay privacy-conscious. Keep an eye on global privacy regulations and make sure tracking methods comply. Remember, trust is key in building long-term customer relationships. And fourth, test and adapt. AI thrives on data, but it's not perfect. Regularly review your campaigns to make sure you're getting the ROI that you're expecting. A/B testing is still your best friend. Remember, AI is a tool. It's powerful, but it's not a silver bullet. Your expertise as a marketer is what brings it all together.
And there you have it, a crash course on how AI is changing the game for conversion tracking in digital advertising. If you found this post helpful, share it with your fellow marketers, and don't forget to subscribe for more insights. As always, thanks for tuning in to Let's Talk Social, where marketing meets the future. Until next time, keep testing, keep learning, and keep connecting. Bye for now.
So this is a big one, guys. You know, post-blog, just casual talking here for a few minutes about today's topic. Conversion tracking is something that marketers really struggle with because it's basically, "Did you succeed or not at your job inside of the ad campaigns?" And so making sure that this stuff is set up is something that's severely overlooked inside of a lot of agency or solo marketer settings when they go to do stuff for clients, whether you're just a freelancer or an actual agency.
And the reason it's so overlooked is really just because it's a hard thing to set up and it's not really easy. It's kind of a pain in the butt to get all of these things talking to each other and to make sure that the attributions are set up properly. And then you also have the argument of what the attribution window is.
So an attribution window is: if they click my ad today, how many days more do I get credit for that sale if it happens on the website or that form fill, whatever it is. So if we're looking at a Facebook ad campaign that's running and we have a one-day attribution window, which is the shortest that you can set up for somebody, they can click it. And then if they go back within 24 hours or in that same traffic period—so they clicked directly from the ad and they just go straight up and buy it right after, or if they wait a whole day, up to one day, and then go back and buy it—then the ad that they clicked on gets credit for that.
The issue is that you can do a 28-day window, and a lot of clients, from what we've seen, don't like the 28-day window because if we're just running ads for them, so they're doing all of the content promotion, they might have a radio ad or a TV ad up, or, you know, insert different ad style here. Those all also play into the thing, but then your ad campaign is trying to take credit for the entire purchase or lead fill or whatever it was.
And so especially when it comes to purchases, if we know the exact amount of product that they bought, so let's say it was a $220 purchase, that typically is going to stack up with all the other purchases inside of the same—like, let's give it a 30-day time period, for instance. And so that can rub a client or business owner wrong because if you say, "Hey, you know, last month we did 30,000 in sales and those are all from our ads," they're going to say—in some cases—they could say, "Hey, we've been doing everything without ads and been doing 20,000 a month every single time, and now you guys are saying you did 30,000."
We did do 35,000 last month, but that's only a $15,000 increase, so there's no way that your ads had a $30,000 increase or like can take credit for $30,000 with the sales. So we had moved our attribution window and all of our ads from the 28-day period down to the one day. So it's as fair as you can humanly possibly get it to make sure that we're trying to take credit, and this sounds kind of weird, but like for as little as we possibly can. But it's also just the most "quote unquote" fair way to do it, where if they see one of her ads, they click it, but then like a week later, they've now subscribed to them or followed them or whatever it is, and then see different content that was purely from the client. We don't get credit for that, even though we brought them in the door from the first place.
So there's a push and pull there. It kind of stinks, you know, not being able to take credit for all of that, but I mean, we've had multiple e-commerce clients that have kind of kicked back and said, "Hey, there's, you know, if we were doing a hundred thousand before we hired you guys, and then we did 110, but you're claiming 40,000," and attribution and like saying that the ads generated that amount. Like, there's no way because it went up on my 10 grand, and you're claiming 40.
And the tough point about that is that we—the person did click on our ad or see it or view it or go into the website from it or whatever it may be, but if they didn't buy that same day or maybe they even did, it's still like someone that could have been a current follower that was going to buy anyways or a prior follower that was going to buy anyways. So there's kind of a very gray area there with how the attribution window is set up and how that works.
But as far as like the AI thing goes, you know, they're talking about data tracking and all of that sort of stuff. You'll hear in the other blog post I released on TikTok and stuff and just the expectation of privacy and data collection targeting is getting harder and harder to nail down. And I've said it in two other blogs, and I'm going to go over it again about like the groups thing with just trying to market into groups instead of just doing the targeting option, but Facebook advertisers, TikTok, LinkedIn, YouTube, whatever it is. Us marketers are going to have a pretty tough time in the coming years, just because they're moving to the "quote unquote" smarter AI stuff, but at the same time, removing more data that people can collect.
So there's going to be this assumption from AI—not necessarily a guarantee—if they're using the AI to do the targeting and all of that stuff. So without data collection and tracking where people are going, you know, their digital fingerprint, so to speak, it's going to be really hard to find people and to make sure that the ads are also being attributed to the proper place because the digital fingerprint right now can track you across from Facebook to TikTok to Google to wherever, and a lot of these companies sell information to each other.
So even if you're on like a news website, if they have the Google ads that pop up on the side of the blog article or whatever it is, they're still relaying your IP address and all of that stuff back to Google so that Google's campaigns can get smarter. But if that connection breaks, or let's say Facebook can't relay the info back over to TikTok, I'm not saying they send each other stuff or sell data to each other, but if those connections start to break, then it's going to be extremely harder to market to people based on the targeting and everything.
And the other blog post I had on just groups and all of that, I think is an interesting topic. And just the next few years of social media is going to be really interesting. So let me know what your thoughts are, what your comments are on this. I would love to hear more people's general reaction to what was said in the post today. If you have like friends or family members that are in the ad space, send this to them because it really could help in terms of the actual targeting and everything.
Or if you have like business owners that are friends or you're a business owner, you know, I give a few tips in here on how to check in on the ads themselves. There's even Google Chrome plugins like BuiltWith. You can go to your own website, open the Chrome extension—again, it's called BuiltWith—and you can see what tags are on your website.
So you can know if the tag is at least there, but also, without going into the ad campaign itself, you're not going to know if tracking is set up properly. So I'm not saying put your agency or advertisers on blast and go grill them on, "Is the setup properly?" But it's definitely something to double-check because now that most of this is done with AI, you can be throwing a lot of money in the wrong place and wasting it, quite frankly, just on the tracking not being set up because the AI doesn't have any information to learn from at that point.
So, hope you guys appreciated it and we'll see you in the next one. Thanks.
‍
