Your Genesys Blog Subscription has been confirmed!
Please add genesys@email.genesys.com to your safe sender list to ensure you receive the weekly blog notifications.
Several years ago, we had a customer who decided to build and deploy a major bot initiative on its own using a simple yet supposedly “amazing” messaging platform. The company was sure it would save a lot of time and money. No help needed from Genesys.
At the time, it seemed like our loss. But four years later — and with no solution in place — that former customer finally admitted it had underestimated what the initiative required. This experience sounds to me like a very costly – and lengthy – encounter with the Dunning-Kruger Effect.
Dunning-Kruger Effect Strikes Contact Centers with Little Warning
A paper published in 1999 launched the Dunning-Kruger Effect. Now widely accepted by psychologists, this theory posits that low performers overestimate their abilities by a wide margin. Largely unaware of just how deficient their expertise is, they suffer with a double burden. Not only does their incomplete knowledge lead to mistakes, but not being inquisitive enough about their choices prevents them from recognizing when they make mistakes or when others make better choices.
If you or someone on your team recommends or influences purchasing decisions in contact center technology, hyper-confidence can have costly consequences.
A Little Knowledge Is a Dangerous Thing
In the case of our customer, the company had some in-house skills and some experience with bots. But there’s also the chance that bad consulting advice drove decision-makers to poor choices. What the company didn’t have was an enterprise view of what the business needed, or deep knowledge of overall contact center technology.
It oversimplified requirements and didn’t recognize critical gaps in building out the ecosystem. Dunning-Kruger was a brutal lesson in contact center decision-making. Here’s how it worked.
- Relying too heavily on peer comparisons
Anecdotal evidence is a strong motivator in forming opinions. And confidence is so highly prized in the tech industry that many people would rather pretend to be skilled or knowledgeable than risk looking inadequate.
Most people think they’re smart, but the Dunning-Kruger Effect can impact anyone. Your IT shop might hear from some peers that they’re running workloads on Amazon Web Services (AWS) or they’re jumping on the Microsoft Azure bandwagon. This is an example of picking projects with no awareness of how the platform supports — or interacts with — other parts of your business, or whether it’s right for your needs. It’s a classic Dunning-Kruger wrong move: Looking at general claims made versus keying in on specifics.
Instead, use objective measures to guide your selection process. This will eliminate up-front flawed estimates based on faulty misconceptions. Analyzing the details makes it harder to maintain the illusion of a solution being better than it really is. For example, what are all the measurements and statistics that should go into your initiative? How should you think about employee performance? How do the solutions you’re building out serve those outcomes you’ve identified as critical? In short, does your plan measure up to the demands of your contact center?
Solicit feedback on the performance of solutions you’re considering from people who aren’t friends or colleagues — or your over-enthusiastic IT guy. Another way to step away from peer comparisons is to compare your contact center performance today against its previous performance. Identify differences and understand why they occurred.
- Skills don’t always transfer to more complex projects
As the co-author of the original paper, David Dunning said, “The trouble with ignorance is that it can feel just like expertise.” Many individuals mistakenly believe their experience and skills in one particular area are transferable to another — without understanding complexities within what appear to be subtle nuances.
Let’s stick with the example of bots. Standalone bots are relatively easy to build and deploy. Businesses can see how a bot is performing in terms of answering a specific question, or the number of self-service interactions the bot enables. It’s a good start but building a bot that works well could give you a false sense of mastery. This is your first Dunning-Kruger red flag.
You’ve experienced only a very narrow view of bots that leaves a lot of potential on the table. But you can’t see it. In fact, when integrating bots into your overall strategy, making a good bot can be extremely difficult.
