Every business owner who has sat through an AI demo has seen the impressive version. The polished use case. The dramatic before-and-after. The capability that sounds like it could change everything. And then they go back to their business and try to figure out where any of that actually applies, and the answer isn’t obvious.

The problem with starting from the impressive version is that impressive isn’t the same as useful, and useful isn’t the same as where AI actually delivers fastest for a growing business. The fastest returns almost always come from the least exciting place: the work that nobody wants to do, that takes longer than it should, and that adds up to a meaningful portion of every workweek without anyone stopping to calculate the cost.

The Insurance Underwriter’s 49 Minutes

Henry Hays tells a story from one of his engagements that makes this concrete. An insurance underwriter at one of his client companies was spending significant time on a specific part of her job: the initial deep dive into a 101-page policy document. Pulling out the relevant data, identifying the key provisions, building the foundation for her actual analysis. It was thorough, necessary work, and it consumed a substantial portion of her day.

After AI was introduced to handle that initial extraction and summary, she got 49 minutes back. Every time. On every policy.

She knew immediately that she was never going back to doing it the old way. Not because 49 minutes is a trivial amount of time — it isn’t — but because the work she was doing in those 49 minutes was the part of her job that required the least of her actual expertise. The analysis, the judgment, the decisions that required a trained underwriter: those stayed with her. The mechanical extraction work didn’t have to.

That’s the pattern worth looking for in your own business.

What Boring, Mundane, and Tedious Actually Looks Like

The framework Hays uses when he works with a new company is to look past the exciting AI applications and focus on three categories of work: boring, mundane, and tedious. The specific tasks vary by industry and by role, but the underlying pattern is consistent.

Boring work is the kind that requires attention but not judgment. Reading through documents to find specific information. Formatting data from one system so it can be used in another. Pulling together information from multiple sources before a meeting or a decision. These tasks have to be done correctly, but they don’t require the expertise of the person doing them. They just require time.

Mundane work is the repeatable stuff that shows up on a schedule. The weekly report that always pulls from the same places and gets formatted the same way. The follow-up emails that follow a predictable pattern. The status updates that require gathering information from several places and synthesizing it into a consistent format. This work isn’t hard, but it’s constant, and the hours add up.

Tedious work is the kind that requires enough concentration to be draining without being intellectually engaging. Data entry. Transcription. Cross-referencing records across systems. Reviewing large volumes of content for specific criteria. The kind of work where mistakes happen not because it’s complicated but because it’s exhausting.

In most growing businesses, these three categories account for a meaningful portion of every knowledge worker’s week. They’re also exactly where AI delivers fastest and most reliably.

Why Business Owners Keep Looking Past This

The reason most companies don’t start here is that starting here doesn’t feel like AI. It doesn’t match the version of AI transformation that gets talked about at conferences or written up in business publications. Automating a weekly report or streamlining document review doesn’t sound like the future. It sounds like a workflow improvement.

But that’s precisely why it works. Workflow improvements are measurable. They deliver results quickly. They don’t require a large technical investment or a long implementation timeline. And they build the organizational confidence and momentum that makes the more ambitious applications possible later.

Hays is direct about this: go after the work that nobody wants to do, and you’ll almost immediately have proof that this is worth continuing. That proof matters internally, because the skeptics on every team are watching the first initiative closely. A quick, clear win that gives someone back meaningful time every day is the most persuasive argument for going further.

How to Find the Right Starting Point in Your Business

The practical question is how to identify which tasks in your specific business fit this profile. There’s a simple way to approach it.

Ask the people on your team, with genuine honesty, what part of their job they would remove if they could. Not the hard parts. Not the parts that require expertise or judgment. The parts that just consume time without requiring much of what makes them good at their jobs.

Most people know the answer immediately. They’ve been doing that task for long enough that they’ve stopped questioning it. It just exists as part of the job, invisible in the way that routine tasks become invisible when you’ve been doing them long enough.

Those are your starting points. The task that a skilled person spends two hours a week on that requires almost none of their actual skill. The process that exists because it always has, not because it’s the best way to get the result. The report that takes half a day to produce and twenty minutes to read.

Once you’ve identified a few of these, the next step is figuring out which ones have the clearest inputs and outputs. AI performs best on tasks where the instructions can be made specific, the inputs are consistent, and the expected output is well-defined. The more ambiguous the task, the more iteration it will take to get reliable results. Start with the tasks that are most clearly defined, and build from there.

What Happens After the First Win

The 49 minutes the underwriter got back didn’t just save her time. It changed how she thought about the rest of her job. Once she had direct experience with what AI could take off her plate, she started seeing other places where the same approach could apply. That’s the compounding effect that makes the boring-mundane-tedious starting point so valuable: it creates practitioners.

A team that has gotten something tangible from AI is a different team than one that has only heard about it. They’ve built a reference point. They understand what it takes to get AI to do useful work, because they’ve seen it happen. And that understanding is what allows the next application to move faster, because the learning curve has already been climbed.

The companies that are getting real returns from AI right now are largely not the ones that started with the most ambitious vision. They’re the ones that started with the most honest assessment of where time was being wasted, applied AI there first, and built from the momentum that created.

Listen to the full podcast episode here.

Related Resources

• The AI Execution Problem: Why Most Growing Businesses Can’t Get Past Experimenting: Finding the right starting point is one piece of a larger execution challenge. This post covers the full picture of why AI stalls in growing businesses and what it takes to build real momentum.
• Stop Treating AI Like Google: What Business Owners Get Wrong from Day One: Getting the most out of AI on repetitive tasks requires knowing how to work with it, not just what to use it for. This post covers the mindset shift that changes how much value you get from every AI interaction.
• How to Evaluate Any AI Investment Before You Spend a Dollar: Once you’ve identified the tasks worth automating, the next question is how to evaluate whether the investment makes sense. This post walks through the framework for making that call before committing budget.
• Who Should Own AI in Your Company When You Don’t Have an IT Department: Someone has to own the process of finding and implementing these wins. This post covers how to identify the right internal owner when you don’t have a dedicated technical team.