AI Can't Staff a Project

Many leaders believe AI has trouble because the data it uses isn’t good enough.

That’s true, but there’s another problem: often, the question itself doesn’t have a clear answer that a system can use.

For example, in consulting, deciding if you can staff a project gets complicated quickly. Skills might be tracked, but not always in the same way. Availability depends on which schedule you believe. Cost and margin only come into play later.

These are real data issues, but the most important factors aren’t stored in any system.

Being right for a role isn’t just about having the skills. Some people can do the job, but they might not be the best choice to work directly with clients. Continuity means someone already knows the client or the codebase. Ramp-up time is almost never tracked, but everyone notices it. Team dynamics and trust also play a big part in whether things succeed.

So teams adjust. They ask coworkers for advice and depend on the person who really understands the situation.

This human element isn’t just a backup plan. It’s actually how things get done.

When AI comes in, it uses bits and pieces from different systems, while people add their own judgment. The AI’s answer might look reasonable, but even small changes in the question can lead to different results, which isn’t what happens with a trusted expert. People end up double-checking and going back to human advice.

So, while data is part of the problem, there’s also a gap between how decisions are made and how, or even if, systems capture that process.

Most teams aren’t getting rid of the person who truly understands how things work.

Instead, they’re just adding another step before reaching out to that person.