CMOs: Adobe & Zoho AI Reshape Martech in 2026

Listen to this article · 12 min listen

The latest AI tools from Adobe and Zoho are changing how we as CMOs build our martech stacks. This is more than just another automation tool that fills in a first name field. We’re talking about AI that can predict which customers are about to buy and then personalize the entire marketing journey for them at a scale we couldn’t do before. These new capabilities mean we have to rethink our old A/B testing frameworks and get serious about integrating our creative AI (like Adobe’s) with our journey AI (like Zoho’s). The real question is how this stuff actually affects campaign numbers and where we decide to put our marketing dollars.

Key Takeaways

  • Adobe’s Firefly Image 3 and Zoho’s Zia for Marketing Hub 2.0 can slash creative production time. We’ve seen early adopters cut their initial design cycles by up to 40% just by getting the AI to generate image variations.
  • Plugging these AIs in requires a real data governance plan, not just a document no one reads, that defines who signs off on AI-generated images, what data you’re allowed to use for training, and how you prevent the AI from producing off-brand content.
  • CMOs need to get their teams trained on writing effective prompts for image generation and interpreting the predictive lead scores these AIs spit out, because that’s the only way to get value from the AI-driven segmentation and modeling.
  • The budget impact is real: we’re seeing money move away from slow, manual A/B tests and toward rapid, AI-powered experimentation, with pilot programs reporting a 15% bump in campaign ROAS as a direct result.
  • You can’t just “turn on the AI” and hope for the best. Success means setting hard KPIs like the conversion lift from AI-generated creative versus stock photos, or the raw reduction in designer hours spent on resizing assets.

Campaign Teardown: “Future-Forward Finance” with Adobe Firefly Image 3

We had a client, a mid-sized financial firm out of Atlanta, Georgia, who needed to launch a digital savings account for Gen Z and young millennial investors. They faced a classic marketing problem: they needed a fresh visual identity to attract that younger crowd without scaring off their existing, older customers, all while sticking to painfully strict brand and compliance guidelines. We decided to build the campaign around Adobe Firefly Image 3 for all the creative, betting that it would let us produce content faster and test more ideas.

Strategy and Objectives

The main goal was simple: get sign-ups for the new “Apex Savings” account. We also wanted to boost brand engagement with the younger target audience and, frankly, spend less time and money on creative production. We bet that AI-generated visuals, which we could tailor to different groups like ‘young entrepreneurs’ or ‘outdoorsy tech workers’, would beat the generic stock photos and expensive custom shoots in both engagement and conversions. Our targets were a Cost Per Lead (CPL) of $15 and a Return on Ad Spend (ROAS) of 250%.

We ran the campaign for 10 weeks, from February to April 2026, on a $150,000 budget. The geo-targeting was tight, focusing on cities like Atlanta, Austin, and Denver where our target demographic lives. In Atlanta, we got even more specific, targeting Instagram and TikTok users in neighborhoods like Midtown and Old Fourth Ward.

Creative Approach with Adobe Firefly Image 3

Our creative team got to work with Adobe Firefly Image 3 (which you can access through Adobe’s AI platform) and started hammering out prompts. Working with the client’s brand book, they wrote detailed descriptions to generate thousands of images showing young people in co-working spaces, enjoying life, or in aspirational settings that hinted at financial success. A typical prompt would be something like: “Photorealistic image of a diverse group of 20-somethings collaborating on a laptop in a modern, sunlit cafe, with subtle financial charts visible on screen, lively colors, shot on a Sony Alpha 7 IV.”

We ended up generating over 5,000 distinct image variations in different styles and aspect ratios. This let us A/B test a huge number of creative concepts. The AI’s knack for spitting out variations of a core prompt, all while keeping the same look and feel, was a lifesaver because it meant we could supply every ad placement with perfectly sized, on-brand creative without starting from scratch. We also used Firefly’s generative fill feature to adapt some of the client’s existing assets for new ad formats, which saved our designers about 25 hours a week of mind-numbing resizing work. When we saw early data showing certain visuals weren’t working, we could pivot almost instantly.

Targeting and Placement

Our targeting was aimed at people aged 18-35 interested in finance, tech, and personal growth. We ran the ads on Meta’s platforms (Facebook and Instagram), TikTok, and across programmatic display networks. We built lookalike audiences from existing customer lists and layered on third-party data. For the Atlanta market, we drew a 5-mile radius around the Georgia Tech and Georgia State University campuses to hit that younger demographic hard.

The volume and diversity of the AI-generated assets really paid off. We saw a 30% higher Click-Through Rate (CTR) on average from the AI images compared to our control group, which used standard stock photography. Some of the best performers were images of people enjoying outdoor activities while quickly checking their finances on a phone. Those hit CTRs of 2.8% on Instagram stories. Because we could swap out duds for new AI options in minutes, our audience never saw the same ad twice, which kept ad fatigue down and engagement up for the whole 10-week run. We ended up with 15 million total impressions, beating our forecast by 15% because the ads were just more relevant.

The final numbers were solid. The campaign hit a CPL of $12.50 (beating our $15 target) and a ROAS of 280% (way past our 250% goal). The better CTR and conversion rates from the AI creative drove this. The final cost per conversion for a new account sign-up came in at $60.

What Didn’t Work

While Firefly was great for images, it struggled to create photorealistic text inside those images, which is a problem when you have complex financial disclaimers to include. That part still needed manual work, slowing things down. We also had to be careful with the initial image outputs. Despite prompting for diversity, the AI sometimes generated faces with subtle distortions or people in unnatural poses, so every image needed a human review before it could go live. Human oversight is still absolutely necessary.

Optimization Steps Taken

To fix the text problem, we changed our process. We started generating the images clean, with no text, and then had designers add the copy as an overlay. It was an extra step, but it ensured we used the right brand fonts and met all the legal requirements. For the weird visual glitches, we put a stricter QA process in place with two designers reviewing every single AI image. We also got better at writing prompts, adding negative keywords like “no distorted faces” or “natural hand gestures” to get better results from the start. This iterative prompting process cut our rejection rate for AI images by about 20% over the life of the campaign.

Strategic Impact of Zoho AI for Marketing Hub 2.0: Enhancing Customer Journeys

At the same time, the client was rolling out Zoho Marketing Hub 2.0 and its Zia AI features to handle leads after they came in. This was an overarching strategic shift that fed directly into our “Future-Forward Finance” campaign’s conversion funnel.

Zia AI for Personalized Engagement

Zia for Marketing Hub 2.0 (which is Zoho’s AI assistant) was put to work analyzing the behavior of every lead we generated. As soon as someone clicked an ad and gave us their email, Zia started tracking their every move on the website and in our app. It was looking for patterns to predict which leads were most likely to actually open an account based on engagement scores, time on page, and what content they looked at.

For example, Zia would automatically flag leads who kept coming back to the “interest rates” page or used the “savings calculator” as high-intent. This let us segment our leads on the fly. Instead of one generic welcome series for everyone, Zia triggered different email flows for different people. A high-intent lead might get an email about specific rate tiers and an invite to a webinar, while a less engaged lead got more general financial literacy content. This is the real difference between proper AI and basic automation, the system is making decisions based on predicted intent, not just following a pre-set ‘if/then’ rule you wrote six months ago.

Results and Observations

The results of integrating Zia were clear: we saw an 18% lift in the conversion rate from lead-to-signup for the group that got the personalized journeys versus the control group on the standard email sequence. Open rates for the Zia-triggered emails were 22% higher, and CTRs were 15% higher. Zia’s predictive models also helped us win back about 5% of leads that had gone cold by telling us the best time and channel to try and re-engage them.

One of the most interesting things we saw was Zia’s ability to spot “churn risks” in the first 30 days after someone signed up. It analyzed how people were using (or not using) their new accounts and could flag those with low engagement. This let our customer success team jump in with a targeted email offering a 1-on-1 setup call or a link to a ‘how to make your first deposit’ video which cut our early churn by 7%.

Challenges and Future Enhancements

Getting Zia up and running took a lot of work upfront, feeding it historical customer data and constantly tweaking the rules for what it defined as a ‘high-intent’ lead. We also had to constantly check that Zia’s personalized emails didn’t accidentally make an income claim or promise a specific return, which meant our legal team was in almost every marketing meeting to ensure we stayed compliant. And while Zia was great at personalizing the journey, the content for that journey still had to be made by humans.

The next step is to see if we can get generative AI tools like Adobe’s to feed images and copy directly into Zia’s recommendations, creating personalized content automatically. These new AI releases are catalysts for a big shift in how marketing teams work. CMOs need to get the budget for the infrastructure to connect these systems. And they have to find or train people who can actually write good prompts and manage the data governance frameworks that keep it all legal and on-brand. Marketing leaders who aren’t testing this stuff now are going to be left behind.

How do Adobe’s new AI features benefit creative teams specifically?

It’s a massive time-saver. Adobe’s AI, and Firefly Image 3 in particular, cuts down so much of the grunt work like resizing images for different platforms or removing backgrounds. This means your designers can test a dozen visual ideas in an afternoon instead of spending all week on one concept, letting them think more about campaign strategy instead of just production. Our team cut its initial design cycle time by an estimated 40% on the finance campaign.

What are the key advantages of integrating AI into customer journey personalization?

You get personalization that actually adapts to the customer in real time. An AI like Zoho’s Zia watches how a lead interacts with your site and can change the marketing sequence on the fly. If they keep looking at the pricing page, the AI sends them a case study on ROI, not the generic welcome email. It’s a big step up from old-school segmentation where everyone in the ‘small business’ bucket gets the same emails, leading to much higher conversion rates, we saw an 18% lift.

What data governance considerations are critical when using AI in marketing?

Data governance is huge. You have to think about privacy, bias, and security. You’ve got to make sure you’re compliant with regulations like GDPR and CCPA, but also set your own internal rules for ethical use, like having a human review process to check for bias in AI-generated images of people. You also need a way to audit the AI’s outputs, like a simple checklist review before any AI content goes live, to catch mistakes and protect the brand.

Can AI tools like Adobe Firefly Image 3 replace human designers entirely?

No, not even close. These AI tools augment human creativity. They’re assistants, not replacements. They’re great at the ‘how’ but useless at the ‘why’. A human designer is still needed for the initial strategy, for writing clever prompts that get good results, and for that final quality check to catch the weird AI artifacts, like people with six fingers, that these tools still produce. Our campaign would have failed without that human review.

What is the most effective way for CMOs to measure the ROI of new AI martech investments?

You measure AI’s ROI with a mix of efficiency and effectiveness metrics. On the efficiency side, you track things like reduced creative production time or a lower CPL. For effectiveness, you look at the big-picture results like increased CTR, higher conversion rates, and a better ROAS. You absolutely have to set a baseline before you start, and running a control group against your AI-powered campaign, like we did, is the only way to prove the tool is actually making you money.

Editorial Team

The Editorial Team at CMO News Time is a group of writers and editors who cover marketing, brand, and growth. They research each story and publish under a shared desk byline — experts in their fields, working as one team.