POCs often feel like a gamble. You're testing the boundaries of your business's technological capacity and your ability to change at the same time. If your POC is an anomaly or you're unprepared to activate it, you risk missing out on technology that could keep your business on top. The choice for a data specific POC is critical, because organizational resistance to change is one of the biggest barriers to data integration. So if your POC fails, users are unlikely to adapt to support later data initiatives.
What to Avoid
Ultimately, when it comes to the evaluation of data science solutions, POCs should prove not just that a solution solves one particular, specific problem, but that the solution in question will provide widespread value to the company: that it’s capable of bringing a data-driven perspective to a range of the business’s strategic objectives. But the choice of which POC is best for the organization and can bring the most value is agonizing.
After working side by side with lots of businesses of all sizes and from all kinds of industries on data science POCs, we've seen it all. But regardless of which POC you decide is the most valuable, there are universal pitfalls. We've compiled a list of the top 6 mistakes to avoid during data science POCs - a crash course, if you will, to ensure you don't fall into one of these common traps (you can see a larger version here):
