I recently had the opportunity to attend the Breaking Silos: Leaning Into Data For True Omnichannel Success conference: Breaking Silos: Leaning Into Data For True Omnichannel Success Hosted by American Express.
While the goal of omnichannel marketing isn’t new, what made this conference thought provoking was the way emerging technologies – and AI – are coming into play.
Here are five big takeaways:
- Having one centralized data hub is key to not only a better customer experience, but also to unlocking the full potential of AI.
According to a recent Forrester study, 70% of marketers struggle with identifying customers across channels. This leads to an inconsistent experience where customers may not get the most relevant offers – or worse, get targeted with something they have already bought! Shout out to Sureshmi (Sue) Shlapakovsky from Audible on a great talk on how they solved this issue – and even added AI tools to the mix.
- Want to create a hyper-personalized experience? Create audiences based on actions rather than attributes.
Marketers will get much better return on investment (ROI) if we stop looking at attributes and start looking at people as individuals and factor their actions into how and what offers we show them. Have they downloaded your app? Do they always use discount codes? How often do they purchase? Are most of their expenses on restaurants?
- Traditional digital advertising is not as effective as it used to be – which is why direct mail is becoming even more important in driving both online and offline conversions.
It’s not surprising to me that in a recent USPS study, direct mail is seeing a 2.5x ROAS as Google ads, Amazon ads, and over 10x ROAS as Facebook ads. The truth is that many of us are so inundated with digital ads that they aren’t as effective as they used to be. In addition, many brands have seen that direct mail can increase customer spend online – and when they cut it, online revenue goes down too.
- Can Large Language Models (LLMs) help improve omnichannel orchestration?
From channel selection, to frequency, cadence and timing, AI can help us reach people at the right time – and place – in their buying cycle. Not to mention, changes in technology mean that we can create thousands of versions of the same email, direct mail piece, ad, etc. if we prepare a bit on the front end.
- To successfully use AI in marketing, you need both high-quality data and a human-insight layer.
There is so much context that LLMs may not have – so it’s important to build in guardrails for usage and ensure you always have a human layer of oversight. For Audible, this meant getting the editorial team involved when building out their AI-powered search tool to ensure book recommendations met distinct needs. For you, this may simply mean having someone always double check what the AI tool outputs.
I’m personally excited to see how we as marketers can keep innovating in this ever-changing world of AI!