5 questions to ask AI vendors before buying a tool

There are many ways to use AI in marketing, and it feels like for every smart initiative, 10 AI vendors have cropped up with a tool to address it.

At the beginning of this wave, I took more calls and answered more emails than I do these days. Over time, I realized I was asking vendors the same handful of questions to assess whether their tools were worth deploying.

If you’re in the same boat and overwhelmed with vendor outreach, here are five questions to help you figure out whether they’re worth your time, along with my rationale for asking them and what I’m looking to hear — or not hear.

This question should help you understand the purpose of the tool and — crucially — whether the value it creates maps to real business outcomes. 

If the vendor can’t clearly state the challenges or use cases the tool addresses, it wasn’t purpose-built to solve a real problem your team faces, whether you’re in-house or at an agency. Beware of vendors who try to dazzle you with feature-heavy language but can’t explain the business benefits those features deliver.

If a vendor identifies at least one existing team problem the tool solves and explains how it improves business outcomes, it’s a good idea to keep talking. A great follow-up question is to ask for a case study showing how the tool was used and the outcomes it delivered for an organization similar to yours in size and vertical.

Look for benefits such as “increases output” or “identifies gaps in tracking to speed up troubleshooting.” But don’t rush to invest in tools that promise to “save time” (even if they truly do) unless you have a plan for how you’ll use that extra time.

The answer to this question should tell you whether the vendor built this tool for advertisers or just at advertisers.

Technical chops matter, but so does understanding how a media buyer actually spends their day. If the vendor doesn’t have personal experience in media buying, they need to explain how they researched the media-buying market and incorporated those insights into the tool.

If they have a shallow understanding or lack expertise, that’s a red flag. It’s fine if a rep doesn’t have that expertise directly, but someone on their team should, and you should have access to that person as soon as possible if you intend to keep talking.

If a vendor has a story or background that led them to identify a problem you can relate to firsthand and decide to build the solution themselves, that’s compelling. A founding mission that addresses your existing challenges is a solid foundation for a tool that can make a difference in your team’s performance.

3. What case studies, real use cases, and results can you share?

I touched on case studies a few paragraphs above, and they’re a must-have in a new, fast-developing industry. I’d be looking to understand whether the vendor has an appealing track record with customers like me or whether we’d be an early adopter.

If you fall into the latter camp, there are pros and cons, just as there are with any advertising beta you can access before your competitors. You may get a leg up by finding an important growth accelerator before your competitors do, you may spin your wheels trying to work through bugs, or you may find that the tool simply doesn’t deliver on its promises.

If you can’t trust a tool or there’s a risk you’ll need to provide detailed feedback to get it working, it might not be the best use of your time and money unless you believe what it could eventually deliver would be a game-changer.

If you’re clearly going to be an early adopter and the vendor isn’t willing to be flexible on contract terms that help mitigate risk, that’s a nonstarter. More established tools will likely deliver more value consistently, although they’ll have less wiggle room on pricing because of it. Newer tools that take a hard line on pricing and contract terms, however, likely won’t be good long-term partners.

For established vendors, you need to see specific, relevant case studies with real numbers from advertisers in a similar space, of a similar size, or with a similar use case.

If they’re early-stage companies, the best answer is honesty: “You’d be one of our first clients in this vertical. Here’s what we’ve seen elsewhere, and here’s what that partnership would look like.” That transparency is a green flag.

---

4. Who owns my data, and how is it being used to train models?

It’s interesting how easily people share data with AI and AI tools in the rush to find a competitive edge. This is something I’d strongly caution potential buyers to consider before signing anything.

Watch for any answer that suggests your data is used to train shared or third-party models without your explicit consent. Another red flag is vague or deflecting answers or terms of service that contradict or muddy what the salesperson tells you verbally.

You own your data, full stop.

The vendor should be able to clearly explain its data-handling practices, including where your data is stored, how long it’s retained, whether it’s used for model training (and, if so, only to refine your own instance), and what happens to your data if you stop using the tool. This needs to be in the contract, not just a verbal assurance. If it’s not there, insist that it be included before you sign.

5. What does implementation actually look like, and what does success require from our team?

Before you commit money, you need to understand the real cost of adopting this tool. That cost includes more than the price. It’s the time, the internal lift (including integration, training, and QA), and any potential disruption to your existing martech stack.

If this requires resources your team doesn’t have, or if you realistically can’t dedicate the time to make good use of the tool, it’s not worth the investment just yet. Many marketers could avoid wasted martech spend if they asked this question and seriously considered the answer.

No tool should be one-size-fits-all, but easy implementation and intuitive design are crucial to getting your team to adopt and stick with it.

Don’t let AI hype rush your decision

I know firsthand that a lot of these tools sound too good to be true — and often, they are. You need to balance growth ambition and curiosity with a bit of caution.

Remember that we’re still in the early stages of AI adoption. If a tool seems too expensive or hard to onboard, or the contract is more rigid than it should be given the tool’s track record, a more attractive solution will likely emerge in the next few months.

When in doubt, ask for a free trial. Assuming it won’t create too much work for your team to integrate the tool, that could be the right next step in determining whether you’ve found your next competitive advantage.