Sage team members Rachel Lord and Rob Zodda recently attended the ANA conference, AI That Performs: Embedding Intelligence into Creative, Direct Mail, and Omnichannel Operations, where they soaked up ideas for shaping smarter, faster creative — from AI-powered workflows to what’s next in direct mail.
Here are five ideas that resonated with them, and what they mean for your marketing.
-
Your customers don’t think in “channels.” Neither should you.
Direct mail (DM) can mistakenly get written off as an expensive legacy channel. But experts and data show that isn’t the case. In fact, direct mail is having a resurgence. DM is often the only part of the customer journey that happens in the physical world, and it does more than drive its own response rate — it has a halo effect that lifts every other channel. This makes matchbacks and multi-touch attribution models key to measuring performance.
-
Before you cut direct mail, know what it costs to bring back.
Chances are this story rings familiar to you too (and it always plays out the same way): leadership cuts direct mail to save money, the business softens, and suddenly everyone wants it back. But DM programs take real time and investment to rebuild — and the vendor discounts and goodwill you’d built up? Gone. The takeaway: scale back when you need to but think twice before cutting it altogether.
-
AI works best when you start small — with real friction points.
A lot of us have felt the pressure of “we need to be using AI.” But the smarter starting question isn’t how do we use AI — it’s where is our process actually breaking down? Start there. A couple of examples shared by presenters:
- Copy translation: Dense product copy can be run through a proprietary AI agent trained on your brand guidelines to help create customer-friendly copy. Then trained writers just need to improve and optimize by using their judgement.
- Idea triage: Twenty rough creative concepts can be fed into an AI agent to stress-test them and generate a hypothesis for each. Then your team can prioritize and build out the strongest ideas for testing.
In both cases, AI does the heavy lifting — but people still make the calls.
-
Let AI catch what you miss. Don’t let it write what only you can say.
“Slop” was Merriam-Webster’s word of the year for 2025, and LinkedIn just launched a button to flag it. We need to remember that our goal as marketers should be to make good work — not just to make more work. AI can’t replicate the empathy and point of view a person brings to creative work. But it’s great at catching what’s missing: checking a revised piece against earlier edits or reviewing copy against your brand voice guidelines. Think of AI as an editor — not an author.
-
Synthetic data is having a moment — here’s why it matters to you.
This one took us a minute to wrap our heads around. Synthetic data isn’t real observed data, and it isn’t the same as anonymized data — it’s artificially generated to mirror real-world patterns. That makes it useful when real data is expensive or hard to come by. It’s especially valuable in regulated industries like financial services. Paired with AI, it opens the door to building microsegments and testing how creative might perform before it ever reaches a real audience.
Here’s the bottom line.
AI is a tool for finding friction and filling gaps — not a shortcut around good creative or channel strategy. That’s exactly the lens we’re bringing to our own work with clients right now, and we’re excited to keep sharing what we learn along the way.
