Houston, we have a benefits engagement problem. In the past several years, HR teams have been scrambling to help employees better choose, use and interact with their benefits. And for good reason — benefits make up 30% of an employer’s total compensation costs, and yet employees don’t seem to be getting the memo:
2 in 5
employees don’t understand the benefits communications they receive
1 in 3
employees have never taken action as a result of their employer’s benefits communications
Is it because employees don’t care about their benefits? No. In fact, 85% of people said benefits are a critical component when they’re deciding whether to accept a job offer. But they’re not always easy to understand. Navigating benefits is a complicated maze, from ever-changing government regulations to plan design updates and endless acronyms and jargon.
So it’s no surprise that benefits pros are turning to technology to help them better engage and educate employees at scale. But as we look for modern solutions to our benefits engagement problem, there’s one common pitfall to be wary of—artificial intelligence.
AI ain’t all it’s cracked up to be
Ahhh, artificial intelligence: it’s the tech industry’s favorite buzzword, and these days it’s helping us do anything from build driverless cars to predict natural disasters. So it’s no surprise that AI has infiltrated the benefits engagement space, too.
Many benefits tech platforms are claiming machine learning as their magic bullet, boasting robust AI-driven tools that can help them predict employee behavior and better educate them on their benefits choices. But can AI-powered benefits engagement tools really do all that they say they can?
Unfortunately, the promise of AI has fallen short over the years, especially when it comes to benefits. Why? Because we’re still very much in the era of “narrow AI:”
“Also known as ‘weak’ AI, narrow AI is the type that exists in our world today. Narrow AI is programmed to perform a single task — whether it’s checking the weather, being able to play chess, or analyzing raw data to write journalistic reports.”
Why is that a problem when it comes to benefits? Because engaging with benefits is not a single task.
An employee’s benefits journey is a tangled web—from enrolling in the right plans to making year-round, in-the-moment decisions that are incredibly personal: like how to find the best in-network provider, whether to schedule a telemedicine or in-person appointment, or which prescription will give them the best bang for their buck.
Employee benefits are a tangled web
The other problem? AI takes time. Just like humans, AI algorithms need time (and a lot of information) in order to learn and get smarter. The more complicated the problem, the harder it is for AI to do its job effectively without a huge depth of data. So those newer tools that claim to make effective benefits recommendations using AI? It’s just not possible. They don’t have enough data to make smart predictions.
Simply put, AI as it exists today isn’t sufficient enough to anticipate employee’s needs and behaviors, predict the best outcomes, and drive them towards greater decision-making.
Avoiding the snake oil
When applied correctly, AI is a powerful tool. Those who wield it can leverage that power to accomplish incredible things with relatively limited human effort. But that power can also be a lure—without proof that it actually works.
Like so many new technologies that came before, AI-powered benefits engagement technology is a shiny object that many may be drawn to, but lack understanding about. The risk? A spread of misinformation and tolls that overpromise and underdeliver.
If you’re thinking about investing in a benefits engagement tool and are struggling to understand how artificial intelligence works, you’re not alone. (After all, your HR plate is full enough—we don’t need to add “machine learning expert” to your long list of responsibilities!)
And for so many applications of AI, you don’t need to understand it to know that it works. You don’t need to know the ins and outs of facial recognition to know that it’s an easy way to open your phone. You don’t need to know how eCommerce sites collect your data in order to have a more seamless and personalized online shopping experience.
But for something so complicated and long-term like navigating benefits and healthcare decisions? How much can you rely on a recommendation provided by artificial intelligence that you don’t understand?
Benefits pros: before you sign on the dotted line, ask your technology providers for evidence. How exactly does the tool they’re selling use AI, and can they prove that it actually works? Does it help employees make smarter benefits decisions than they would have otherwise?
Putting the human back in technology
As employees resign at record rates and demand sweeping changes to how and when we work together, company leaders are scrambling to form more human connections with their workforces and better communicate their employer value proposition.
A huge part of that value proposition? Your benefits package—a strong indicator of how you care (or don’t care) for your employees’ health and well-being.
So it’s funny that so often, we turn to technology to solve these very human needs. Can artificial intelligence make employees feel heard when they’re diagnosed with diabetes, and aren’t sure what resources are available to them? Can machine learning lend an ear when employees are confused about where to bring their child when they break a leg?
Your employees aren’t robots. And they shouldn’t be making healthcare decisions like one, either. Yes, technology can do amazing things to help us educate employees and support and supplement the human connections we make at work. But if it’s cold and inhuman, and its only “secret sauce” is a shiny new industry buzzword? That’s not useful for anyone.
Let’s give employees a benefits tool that actually helps them. That builds a two-way street between employees and employers. That learns as much about employees as it educates them. That provides truly personalized guidance about incredibly complex issues, in the moment when it’s needed most.
It’s the future we’re building here at Jellyvision, and we hope you’ll join us.