Most AI investments fail for a handful of the same reasons.
That’s the takeaway from Anna Metsäranta, Head of Sustainable AI at Solita, a technology, design and strategy consultancy, in her talk at the Marketing AI Conference (MAICON) 2021.
Here are the big reasons that businesses run into trouble with AI investments…
PS - Have you heard about the world’s leading marketing AI conference? Click here to see the incredible programming planned for MAICON 2022.
The Top Reasons for the Failure of AI Investments
Progress is not always linear in AI initiatives, says Metsäranta. There are lots of stops and starts, changes of directions, and lots of stress along the way. That’s natural.
But a handful of things should be avoided if you want to stand the best chance of success…
- Lack of vision. Sometimes, companies don’t have a clear picture of what they’re trying to achieve with AI, leaving them adrift when making AI investments.
- Unrealistic expectations. The opposite of lack of vision can also lead to failure: having a vision of AI outcomes that just isn’t realistic.
- Data issues. Primarily, data quality. If the data isn’t good, the model, even if successfully built, isn’t going to be reliable.
- Working in silos. Teams need to share actionable information during AI investments. Silos prevent that from happening.
- Lack of human insight. You need to spend a lot of time thinking about the people who will be using the system, and identify failure points along that user path.
- Lack of business design. Even the best AI system won’t generate value unless you’ve taken the time to plan for how it’ll be used in a way that’s meaningful to the business.
- Neglected AI. Once implemented, AI can’t just be put into action and left there. Data is always going to feed into the system, and the system will need to be adjusted. Humans will, at some point, need to determine when the model needs to be retrained.
The Best Ways to Avoid Failure
It’s not all doom and gloom, though. Metsäranta recommends these conscious actions to increase the likelihood of success:
- Communicate a vision that supports your strategy and values.
- Get to know your real data and its possibilities.
- Invest in making your data accessible and reliable.
- Allow time and iterations to find the best algorithms—they’re like recipes and you have to experiment.
- Work iteratively in cross-functional teams.
- Design and manage business changes required by successful AI adoption.
- Secure leadership support every step of the way.
- Manage the cultural change to enable human and AI collaboration rather than competition.
PS - Have you heard about the world’s leading marketing AI conference? Click here to see the incredible programming planned for MAICON 2022.
Mike Kaput
As Chief Content Officer, Mike Kaput uses content marketing, marketing strategy, and marketing technology to grow and scale traffic, leads, and revenue for Marketing AI Institute. Mike is the co-author of Marketing Artificial Intelligence: AI, Marketing and the Future of Business (Matt Holt Books, 2022). See Mike's full bio.