Beta

Testing digital advertising for B2B SaaS

Best practices and must-test variables for B2B SaaS brands looking to boost ad performance
Zach Reiff

B2B SaaS marketers have a particularly difficult challenge in finding ad creative that converts. You have multiple buyer personas, each of which has different pain points and uses software in different ways. And you often have a complex product that requires a bit of workflow retooling to fit into a buyer’s current process. Because of these factors, pinpointing the right ad for the right audience can become a grueling task if done manually. And A/B testing alone no longer cuts it in terms of actionable data.

Automated multivariate testing (MVT) is the fastest route to finding not only which ads work best for B2B SaaS, but why. This is because you’re not just testing ads, you’re testing each individual creative element within each ad — all at once.

As client services manager over Marpipe’s B2B SaaS customers, I’ve overseen hundreds of digital ad creative tests first-hand. Here are the best practices and creative elements that make the biggest difference when it comes to boosting ad performance for SaaS brands.

Personalize your tests toward each buyer persona

Our customers have found success in treating different audiences just as they are — different. Whether you’re a marketing whiz, a product guru, or an engineering aficionado, you deserve personalized experiences. And whether it’s messaging or imagery, everyone has unique preferences.

With the power of multivariate testing, we’re able to narrow in on which audience prefers any given creative, quickly. This allows us to focus on what each individual cares about and also enables the versatility of a SaaS product or service to truly shine.

To successfully personalize your multivariate test, you need to know your audiences’ needs and pain points, and how they would use your product in their specific role. (Someone in marketing will use your software differently than someone in product, who will use it differently than someone in engineering, and so on.) 

From there, relay those things within the messaging and imagery you intend to test. For instance, you might feature a different screen or UI for each audience, depending on which features those personas are known to use most. Or you might test different ways to describe what your product does for that persona, or how it fits into their workflow.

What kinds of ad creative does each role respond to?

Through our customers' tests, we’ve seen that marketers respond well to brand logos of those companies already using the software. They also respond well to images depicting how software affects a customer’s experience.

Product managers and engineers, on the other hand, respond best to images that show how the software works, where it fits into a workflow, featured integrations, and abstractions of UI and the backend.

Deeply understanding these roles and how they use your software will help you understand which creative assets might resonate best. From there, testing is key to proving — or disproving — your hypotheses. 

What if multiple personas like the same ad creative?

If two or more target audiences respond well to one ad or creative asset, you’ve hit the jackpot. This kind of overlap means that it might not be as important to focus on completely different imagery and messaging for these two roles. 

You can scale these winning ads and assets for both or all of these audiences, saving you time by not having to give each role a completely different ad experience.

Two must-test assets: CTAs and UI imagery

We’ve seen our B2B SaaS customers experience huge performance boosts by finding the right CTA and UI image for each of their audiences. Here are some insights into what to test in both camps.

CTAs

The action you ask your potential customers to take can have a huge impact on your ad performance. In a recent B2B SaaS ad test, “Sign up” outperformed “Book now” across multiple buyer personas. 

Be sure to test multiple CTA variants — both in your ad creative and within the social media platform — with varying degrees of urgency. You can also use the information gleaned from CTA tests to inform CTAs found on your website.

UI imagery

It’s common for SaaS companies to use screenshots or product abstractions of their software in their ads. They give people a sense of how the product works and what the user interface involves.

One of the more impactful tests we’ve seen here is testing light mode vs. dark mode UI imagery. For our SaaS customers, images depicted in light mode worked better for engagement, driving a 38% increase in CTR. On the flip side, images depicted in dark mode drove higher conversion rates, resulting in a 32% lower CPA.

Animate your winning static ads to test video

Here’s a trick our customers have recently seen boost CTR by as much as 5x in testing — using basically the same ad creative.

They took their top three static image ads and animated them, turning their static ads into video ads to be tested

This works for a number of reasons. First is that most social platforms prioritize video in their algorithms. The second is that some simple animation can be the thing that really drives home a message. For instance, it might help visualize the steps of your software’s workflow faster than a static image.

Turns out, just adding that bit of movement to your static assets can get you the KPI boost you need.

One thing to look for when analyzing video ad performance against original static ad performance is that not every audience will respond best to video. Some may prefer your static image ads — an insight you can’t know until you test.

In a customer’s recent test, marketers actually favored the static image ads tested over the video ads. These kinds of testing insights should influence how you divvy up your evergreen ads budget — putting spend around the right audiences and the right creative.

Consider the journey of your test

If you take the time and effort to parse out your target audiences, and messages to those audiences, where they go next is really important. 

While planning your ad test, you should also consider where those ads will lead. Try giving each audience its own custom landing page, complete with copy and imagery derived from your ad test. This gives potential buyers the most consistent journey across the board.

Even better — as certain pieces of copy and imagery rise to the top in your test, make sure they appear even earlier in your landing page experience.

To get the right ads in front of the right buyers faster, get on the MVT train.

The big brands of tomorrow are today’s up-and-coming B2B SaaS companies — those making our work lives easier, day in and day out. Fast iteration and creative testing are imperative for conversion- and growth-driving ads. And the fastest way to get there is with modern, multivariate testing.

We love seeing B2B SaaS companies like Typeform and Segment by Twilio experience that a-ha moment after running their first test. Reach out to us with any questions (just hit up support@marpipe.com) or schedule a demo.

We can help you test your way to top ad performance, too.

How to Run a Multivariate Test

The Beginner's Guide

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Tiffany Johnson Headshot

How to Run a Multivariate Test
The Beginner's Guide

Plus, Get our Weekly
Experimentation newsletter!

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.