September 21, 2026 · Jess Pischel
I sat down with a client recently who had three vendor demos booked, a team member pushing a new tool she'd found on LinkedIn, and ten browser tabs open comparing AI platforms. When I asked what she wanted to fix first, she paused. That's the normal starting point right now: plenty of options, no way to sort them.
Here's the filter I use with every client before we touch a single tool. Three questions, in order.
Is the task repetitive and rule based, not judgment based? If it requires a person to weigh context and make a call, save it for later, once your team trusts the tool with the easy stuff first.
Does it sit on a bottleneck your team already complains about? You don't need a new problem. You need the one people already mention in every one-on-one and never gets fixed.
Can you measure the before and after with a number you already track? Hours saved, dollars, response time, whatever your team already has a number for. If you have to invent a new metric to prove it worked, you picked the wrong pilot.
If a task clears all three, that's where you start. Everything else waits.
Here's what that looked like recently. A senior living CEO told me he doesn't find out his team hit overtime until the bill lands on his desk. His team was watching the numbers, just not in one place. Hours sat in payroll, census sat in one system, payer mix in another, and pulling them together every week took time nobody had. By the time anyone noticed the pattern, the money was already spent.
That task cleared all three filters. It was rule based: pull the numbers, don't interpret them yet. It sat on a bottleneck everyone already complained about. And the number was one he already tracked, overtime dollars. We're building him one dashboard that pulls hours, census, and payer mix into a single view his leadership team can check before overtime becomes a surprise, instead of after.
That's most of the AI work worth doing right now. Moving the numbers your business already has into a place where someone can act on them in time to matter.
One thing to try this week: Write down the three tasks eating the most time on your team right now. Run each one through the filter above. Whichever one clears all three, that's your pilot. Don't start with the other two yet, even if they're more exciting. Small pilot, real number, then decide what's next.
If you want a second set of eyes on which task to pick, that's exactly the kind of conversation a discovery call is for. No pitch, just a plain look at where you'd actually start.
Book a Discovery Call →