How to Find AI Value in Your Business: Start With Your People

Many executives I talk with have an uneasy sense that AI could make their business more valuable. There must be time they could recover, costs they could avoid or revenue opportunities they could pursue. What they often lack is a method for finding out where.

They know their business. They do not necessarily know what AI can do with it. That is where The Endurance Group comes in.

I founded TEG in 1998. After years working inside businesses, particularly sales and marketing operations, I have learned that the useful answers usually start with a conversation with the people doing the work.

Start with your people, not a list of AI features

A feature demo cannot tell you where your business is losing time. Your employees can. We ask them to walk us through real work, including the steps that happen outside the documented process.

• What do you do repeatedly each week?
• Where do you get stuck waiting for information or another person?
• What gets copied, checked or reconciled by hand?
• What would you do for customers if you had more time?
• What is growing faster than your team can handle?

That fourth question matters. AI value is not limited to doing today’s work faster. Sometimes it is making room for work that never gets done: better follow-up, deeper customer analysis or a service the company could sell but cannot currently deliver economically.

A practical method for uncovering AI opportunities

1. Choose one team. Sales, finance or another business function is enough to start. You do not need a company-wide AI plan.
2. Map the actual work. Talk with employees, review examples and identify the systems, handoffs and decisions involved.
3. Separate the opportunity from the technology. Define the business improvement first. Then evaluate whether Claude, an integration, a workflow or another approach can deliver it.
4. Estimate value and feasibility. Use operating data and conservative assumptions. Check data access, accuracy requirements and where people must review outputs.
5. Prioritize the builds. Compare expected value, fixed build price, implementation complexity and adoption requirements.

At TEG, the first AI Value Assessment is free. The initial call selects the team. We then spend up to five hours on meetings and research, with an AI Value Report delivered within one week after the assessment is complete.

The report describes recommended builds, what they would do, the process they improve, expected 12-month economic value and fixed implementation prices. You can view a sample AI Value Report to see the format.

AI value includes capacity, avoided costs and new revenue

These are different sources of value. Treating all of them as “savings” can make a business case look better than it is.

• Time freed: a proposal workflow gives salespeople more capacity for customers. It becomes cash savings only if an actual cost falls.
• Hiring avoided: the existing team can handle growth without a genuinely planned additional role.
• Costs removed: a reconciliation workflow allows the company to reduce an outside provider’s bill.
• New revenue: customer reporting becomes a paid service. Value should reflect contribution after incremental delivery costs, not just gross revenue.

We look for overlap, too. The same recovered hours should not be counted once as employee capacity and again as the entire value of an avoided hire.

That is our approach to value-based AI pricing. We agree on the expected value and the price before building. We do not price implementation by the number of hours it takes us.

Build around the business, then teach the team to use it

The world has changed. Buying a software product is no longer the whole answer. Companies can build Claude capabilities around their own processes, knowledge and customers, often inside systems they already use.

Sometimes that means a custom AI agent. Sometimes it means Claude Skills, Projects, integrations, automated workflows or an application. The business opportunity determines the approach.

But a working tool is only part of the job. Employees need to know how to use it, what to check, when to involve a person and what to do when the result is wrong. We train around their actual work and help them recognize new opportunities as their confidence grows.

For example, a salesperson who learns to review a Claude-generated proposal may start asking whether the same information could improve account planning. That is the kind of practical thinking we want to encourage.

Your business changes. Your Claude environment should evolve.

Employees discover new uses as they learn. Claude gains capabilities. Your systems, customers and market change. A solution that made sense at launch needs ongoing assessment and refinement.

AI can create a competitive advantage. Poorly reviewed outputs, inappropriate data access or unreliable workflows can also create costly mistakes. That is why improvement includes testing, clear permissions, human review and attention to what employees actually use.

Managed Claude is our standard ongoing relationship, starting at $1,500/month. We maintain and improve existing workflows, support and train employees, and assess new opportunities. Monthly value reviews help us decide what is working and what should change.

New builds are separately scoped and value-priced. Clients pay Claude subscriptions, third-party software and API charges separately. We agree any change to the monthly fee before it takes effect.

My goal is for a client’s Claude environment to become more valuable as the business evolves. Finding the value is where we start. Continuing to find it is why we stay.

Frequently asked questions

How do you identify AI opportunities in a business?

Start with one team and interview the people doing the work. Map recurring tasks, bottlenecks and unmet opportunities, then estimate the 12-month economic value and assess feasibility, data access and human review requirements before choosing a build.

How do you calculate the business value of AI?

Estimate the value of time freed, genuinely avoided hiring or vendor costs, and additional revenue contribution. Use operating data and conservative assumptions, account for software and delivery costs, and avoid counting overlapping benefits twice. Time freed is capacity value unless an actual cash cost falls.

Does TEG price AI implementation by the hour?

No. TEG agrees the expected 12-month economic value and a project-specific percentage with the client, then sets a fixed implementation price before work starts. The percentage varies by project. The price does not change with actual subsequent results.

What happens after an AI workflow launches?

TEG trains employees, supports adoption and stays as the ongoing Claude team through Managed Claude, starting at $1,500/month. Existing workflows are maintained and improved, and new opportunities are assessed. New builds are separately scoped and value-priced.

What does the free AI Value Assessment include?

The first call selects one team, such as sales or finance. TEG then spends up to five hours on meetings and research. Within one week after the assessment is complete, the client receives an AI Value Report with prioritized recommended builds, expected 12-month economic value and fixed implementation prices. There is no obligation to proceed.