BI Agent

Data > Insights > Action, have it your way.

Many moons ago in a galaxy far, far away we advised the Boards of the Fortune 1000 on metrics, goal setting, and incentive compensation design.

Said more plainly, we determined how much money you were going to earn based on how well you improved your business.

We’d tie Revenue Growth, Return on Equity, Return on Assets, Total Shareholder Return to the board, management team, sales teams, and all employees bonus and equity award payouts.

As you might imagine, we had a fully engaged audience. The one core principle you learn when designing incentive compensation plans is: it turns humans into machines. Be careful what you measure, and pay on, because you’re going to get exactly that.

This “paperclip” problem has been around for longer than most in the AI industry have given credit for. But I digress…

We’re now in the age of Business Intelligence Agents (BI Agents).

What’s a BI Agent?

We get all the company’s operational data in a database, put a human-readable name over the top of each metric (the kids call this a semantic layer), slap an AI chatbot over the top of that, and hand it out to all employees to use.

Why do this? Because in most organizations, people are flying the plane blind. Things are happening in their business in real time but the employees can’t see it. There’s a loop of about a month before they see some of the data show up in a custom slide that somebody else had to create. The leading organizations have these nifty tools called BI Dashboards (sometimes from 2 or 3 different software providers), which become a mind-numbing amount of charts, graphs, slides, and notifications. Then it takes hours just trying to make sense of it. By the time it reaches the senior executives and the boards, the thing that was happening in the business is well in the rear view.

Imagine you’re driving a car, but it takes a month for you to see out the windshield for what’s coming at you. You would have crashed that car in the first few seconds.

Now we have this newfangled thing called LLMs that promise to make this whole process faster and easier. But once you try doing it, you realize there’s issues: a lot of nuance, technical decisions, security, permission, infrastructure, pre-production and production environments, beta rollouts, training, language issues, data store copying and timing, accuracy, and on and on.

So you go to brand names like Palantir or Accenture to do it for you. Holy jeez, look at that bill. How is anyone other than the F50 going to afford that and have the time to build it when we have to prove ROI at every point in time (i.e., we don’t have the luxury of a J-Curve).

Now you’re up sh*ts creek. You have a real business need. You’re behind others who are already operating with this in production and have much faster cycle times. The data you can see is starting to show early indications of changing market share, which is not in your favor.

So what do you do?

You consider the big boys again but you just can’t afford it so that’s not an option. You get bombarded by LinkedIn, email, and phone spam messages trying to sell you something but it’s a brand you’ve never heard of filled with people you have no clue how good they are. This is sensitive stuff.

While your job may not be on the line, you don’t have the emotional energy to manage yet another sideways project. You’re already in firefighting mode with business as usual. Wouldn’t it be nice to take a real Friday afternoon off without worrying about the business?

There’s only one option left: you have to go find the people who have done this before in large, medium, and small operating environments, but don’t charge the same fees. Searching Google and ChatGPT doesn’t give you those search results though because that data doesn’t exist on the open internet. Exhausting.

So now what…

We’ve been down this road before if you haven’t already guessed. Scar tissue, achieved.

Linking compensation to BI Agents is the future.

Let’s assume you figure out the problem of building and deploying the BI Agent to all employees (and bonus points: partners) around the world. Now you can link employee, executive, and board compensation to the metrics being tracked by the BI Agent.

Once every employee knows exactly what they need to do to move the metric(s) they’re responsible for in the right direction, then the business starts working more efficiently and positive results begin to compound.

It doesn’t require layoffs. It just requires getting that real-time data intelligence deployed across the organization, and operationalized using incentive compensation.

Then, and only then, do you begin compounding growth for the company while simultaneously lowering the company-wide stress level.

Easy peasy.

:)

—Sean