Ep. 120 AI-Powered Digital Marketing Ads for Banks & Credit Unions

Artificial intelligence has quickly transformed digital advertising. In this episode of the Hit Record Podcast, Meredith Olmstead and Ida Burr revisit a popular AI discussion and highlight key changes from the past two years. They focus on how banks and credit unions can effectively use AI for reporting, campaign optimization, compliance, and conversion tracking, while keeping strategic decisions in human hands.

Key Takeaways:

• 1.  AI helps you work smarter, not harder. AI can quickly analyze campaign data, identify trends, and surface insights so marketers can spend less time on reporting and more time improving results.

• 2.  Better data leads to better ads.  Strong reporting and conversion tracking help AI optimize campaigns, improve targeting, and drive more meaningful results over time

• 3. People still power great marketing. AI is a powerful tool, but strategy, creativity, and human expertise are what turn good digital ads into great ones.

Transcription:

Meredith Olmstead:

Hi there. I'm Meredith Olmstead, CEO and founder of FI GROW Solutions. We are a digital marketing and sales and service consulting agency and we work exclusively with banks and credit unions. I am here with our digital ads, our director of digital ads or digital ads director, Ida Burr. Say hi, Ida.

Ida Burr:

Hi, everyone.

Meredith Olmstead:

So, Ida and I are revisiting an old podcast of ours. We were actually looking at one of our more popular podcasts, episode 61, and it was all about how to use AI to create digital ads for your financial institution that really convert basically. And so, it was a very popular topic, but then when we looked at it, we realized that we recorded it two and a half years ago. And I was like, how have we gotten two and a half years into AI already? It seems like AI is brand new.

So, I said, Ida, maybe we should revisit this topic and record another podcast on this to make sure that we're up to speed or we're up-to-date on this topic. So, we decided to hit record and really talk about AI for creating digital ads.

So, in the original episode we talked about, we basically were talking about that it is really good for ideas and a starting point using AI is really good for coming up with ideas for keyword lists, for new areas and markets, for titles and descriptions and all those kinds of things.

You can get a lot of really good starting point ideas from AI tools out there, ChatGPT, Perplexity, Claude, all of those things. So, I think back in '24, January of '24, Ida, you were still using, you called it Bard, which was a different AI tool. What AI tools are you now using to help with your digital ad services basically?

How AI Is Changing Digital Marketing Ads

Ida Burr:

So, I've officially made the switch. I pay for ChatGPT. I always get the letters mixed up, but that one has been the one I've been using for a while now and I've experimented with a lot, but that one seems to be the one that's most interactive.

Meredith Olmstead:

Okay, got it. And so, we are using that one, the paid version across the team for sure. Although I know a few of us, our team are also using a couple of other LLMs. So, besides just coming up with ideas and generating... so generating ideas and generating popular keyword lists and reworking content, shortening text to make sure that it fits say within the character limits, what else are you now using AI for?

Because I think that was a lot of what we talked about before and that it's never going to be giving you a finished product that it's really important to make sure you're looking at AI results from with a strategic lens as somebody who has worked in financial services and been doing ads, digital ads for banks and credit unions for a decade. So, what else are you doing with AI that's useful to share?

Ida Burr:

So, one thing that I've been experimenting with is just getting help identifying trends. So, if I see costs go up or costs go down either in pay-per-click or in social, I'll send a screenshot of the results and just ask, "What do you think happened in the past week or what kind of other things should I be looking for?"

And it will actually have me go in and build a custom report and ask me to export it and then drop that into the chat so it can have a really detailed information of, it'll ask sometimes for the time of day that results are happening, the locations that results are happening and it is able to quickly analyze that and find things that would probably take me a half hour to dig through all that to find.

Meredith Olmstead:

Got it. So, it's helping with reporting and basically trying to dig through large amounts of data or results. So, you can download a whole spreadsheet worth of information that's got really specific ad results in it, upload that to the chat tool and it will slice and dice and give you some patterns that might be useful for either reporting or is it useful to make changes? Would it tell you, "Hey, you need to up your budget or you need to run ads like this more," or what kind of changes are you making based on those results?

Ida Burr:

So, sometimes it'll pull something as specific as between 9:00 and 10:00 AM seem to be the most valuable clicks that these ads get and it would suggest maybe changing your bidding a little bit for those times or excluding certain times that you don't want ads to show up-

Meredith Olmstead:

Oh, nice. Okay.

Ida Burr:

... or even locations and things like that.

Meredith Olmstead:

Yeah. Okay. So, it helps you drill down on some of your targeting and the way potentially you might have an ad campaign set up in a certain area based on recent past performance. That's pretty cool. So, you could maximize people's budgets a little bit better for time of day or for ad type based on that kind of analysis. That's pretty cool. What else are you using it for? I know you said something about coding. What do you mean coding?

Ida Burr:

Yes. So, it has really come a long way with coding. So, we like to set up tracking for applications and sometimes, we don't get instructions to set up that tracking. I'll go through each page and try to figure out the best way to track the way that that happens, which isn't always easy because there's not always a URL update or page path flow you could follow very closely. So, sometimes it's button clicks, sometimes it's specific calls to action. But one time in particular, I could not find anything by myself and I thought, let me just see if they have any ideas.

And on the last page of the application, the headline was a specific, thank you for completing your application type thing. So, I figured out a way with chat how to create a conversion based off of that headline existing on a page.

Meredith Olmstead:

Oh, wow.

Ida Burr:

So, then I didn't have to depend-

Meredith Olmstead:

So, it can track the language on the thank you page or the thank you interaction. And then when that shows up, it would mark it as a completed application.

Ida Burr:

Yes.

Meredith Olmstead:

Oh, that's super cool.

Ida Burr:

Because there's not always dependable button clicks or URL changes, so you have to get creative.

Using AI to Improve Digital Marketing Ads

Meredith Olmstead:

Yeah, it's funny because we work with so many different clients that have different LOS systems and online application systems. And so, there's such a variety. I mean, some of them are tried and true. We work with a lot on MeridianLink, but then you'll have some of these one-off random, smaller little application programs that you end up having to troubleshoot your way through the tracking because they're still a little behind the times potentially when it comes to giving access or tracking those fully completed applications. So, the other thing besides just data feeding it in, so I know that chat can also be really good for putting other kinds of data in.

I would assume when you're writing ad copy, you're able to tell chat, "Okay, here's the compliance requirements for these ads, make sure that anything you write meets those requirements." Are you able to do that?

Ida Burr:

Yes. So, making sure there's asterisks where they need to be, that the disclaimer is in the correct spot with it being one click away, things like that for sure.

Meredith Olmstead:

Got it. So, are there other policies and procedures and new features and things that you're able to introduce or understand better or work through with AI?

Ida Burr:

Yes. Well, so if anybody who works with pay-per-click or Facebook, we get emails constantly from those platforms, new policy updates. And sometimes I'll just take an entire email and paste it in and say, can you pull out what would be important for my clients that are in financial industries that are in the United States? Because sometimes, it's like this is only if you live in UK or if you're advertising to Mexico. So, it helps me take out the noise of what are in these long new policies. But also if I have trouble understanding one piece, I'll say, what's one step back?

What should I look into so that I understand what this is referring to type of situation. So, sometimes, I just get a little lost in all the language and it helps me dissect it and take a step back.

Meredith Olmstead:

Well, and I assume that that is going to help with compliance and errors because on occasion you'll get an ad campaign or something that gets rejected. And so, being able to filter through those errors or any of those policies and procedures to figure out, okay, where was it that we missed something in the ad platform or in the ad creation? I mean, human error is a thing. So, occasionally, something might slip through that you've missed and then potentially the AI tool could help find that a little faster for you in a lengthy set of policies and procedures. Awesome.

So, would you say that it's still these tools, I mean, they're continuing to evolve and make your life a little bit more efficient and easy and make ads more successful?

Ida Burr:

Yeah, I think they're definitely getting smarter. There was a while I felt like it started making up things and that's something you always have to look out for on the bottom of every chat that says chat makes errors, please fact check. But it definitely seems to be evolving, especially when we look back two years ago.

Meredith Olmstead:

Yeah, yeah. It's funny. It's funny. It's funny because when we first recorded this podcast, it was like brand new days, but it's funny to think that even a year or two and a half years later, AI is still such a hot topic and it's still evolving and getting better in just this specific area of creating digital ads for financial institutions. So, awesome. Well, thank you so much for the update, Ida. This is super useful. If you're interested in learning more about our digital ad services, please visit us at figro.com. We would love to connect with you.

We also have lots of other great podcasts, blogs, case studies, so lots of good resources for you on our website and otherwise, we'll all just get out there and let's make it happen.