From frustration to excitement: Avoiding AI pitfalls
Your company just completed a major milestone. You have implemented a new AI driven onboarding system. You test it out, and for a moment the feedback is electric. Then, without warning, reality kicks in. Users signing up are more disappointed in the product than before this launch.
What happened? Before the new AI feature, the original product was legacy, but people that signed up enjoyed it. They understood its purpose and found value in it. Now users are dropping faster than flies.
Read on to know how to plan for AI strategies.
The pitfalls of starting with AI experiences
The most destructive mistake you can make when building a product is to start users on an emotional high. As soon as they hit the first complex workflow or legacy systems that's holding on to a thread, they will drop.
This problem becomes worse when you front load your product experience with an advanced AI onboarding system, only to bring users back into older legacy systems. As the saying goes, the higher you go, the harder you fall.
The above chart is a simplified user satisfaction chart for representation purposes. The solid green line represents the original journey flow before AI was implemented, and the dip is a key friction point for users. However, for users that make it past that point, their satisfaction level nearly recovers.
The new AI onboarding starts users at a highly sastified level, however the onboarding stops before this complex point of friction. The perceived change in complexity from AI to legacy will create a stronger emotional response from the new onboarding flow, creating higher drop-off rates.
We all know that onboarding is the most important part to hook users and reel them in. But we need to tread lightly in the world of AI. It's now more important than ever to manage the emotional journey of users as the expectations of product experiences continue to rise.
It may even appear to make sense to implement AI in small doses where there are friction in the journey. However, this can be costly and time consuming if we aren't looking at the entire system. To plan for AI needs to come at the key value points of the product, wherever that may be, and work towards onboarding from there.
Right to left approach
If we map our hero journey, in a linear chart, again this may be grossly simplified - we can work from the key point of value, being the right side of our chart towards onboarding. This way, as the users use the product, their emotional state can only trend upwards. And who knows, given time, your users may eventually just be floating on cloud nine.
From showpiece to value
In the fast paced world we live in today, it's understandable that your business may want to rush the delivery of AI features, but hear me out. Customers don't care if you use AI or static pages, as long as the job gets done and it's done well.
Your role as a product owner is not to ship what is trending. It's to ship what matters most to users. Spend the time to map the user journey, gather feedback, and identify the most crucial pain points. Then improve them. If that requires AI, great. If not, use what technology makes sense.
Even if marketing and business development teams want AI launched quickly, I would argue that having a long-lasting campaign has greater impact in the market. Telling your story and sharing the journey as your team works through the challenges of implementing AI is more interesting than hearing that it's been implemented like everyone else.
Remember, AI is not patch for a bad experience. Like any technology, it is a means to an end.
