Where to start with AI when you have a small team

You don’t need a data team or a big budget to get value from AI. You need one well-chosen task, a few weeks, and a way to tell whether it worked.

Start with the work, not the tool

Most first AI projects stall because they start with a product demo rather than a problem. Instead, ask your team a simple question: what do you do every week that feels like copying, sorting or retyping? Those are the tasks AI handles well.

Write each one down with a rough sense of how often it happens and how long it takes. You’ll usually end up with a list of ten or so. That list is your starting point.

Pick one task with three qualities

• It happens often. Weekly beats yearly. Small savings add up when they repeat.
• Mistakes are easy to catch. Drafting a first version that a person reviews is a much safer start than anything sent straight to a client or donor.
• Someone owns it. The person who does the task today should be part of trying the new way.

Use what you already pay for

Many tools you already use, such as your email, office suite or CRM, now include AI features. Turning on and learning one of those is often cheaper and faster than buying something new. Check what’s included before you shop.

Agree how you’ll know it worked

Before you start, decide what “better” looks like. Time saved per week, fewer errors, faster replies: pick one or two measures and note where you are today. After a month, look again. If it’s working, expand it. If not, you’ve learned something cheaply and can try the next task on your list.

Set a few ground rules

• Decide which information should never go into an AI tool, such as donor, client or health details, unless the tool is approved for it.
• Keep a person in the loop for anything that leaves the building.
• Write down what you tried and what you learned, so the next project starts further ahead.

What would you do with an extra five hours a week? That’s a good question to keep in mind as you choose.