The Rise of the Business Algorithm

We spent months on it.

A team of strategists, designers, and technologists. Cultural immersion. Stakeholder interviews. A competitive landscape mapped down to the seams. Qualitative research, then validation on top of it. A stack of concepts for the future of connected cars and mobility, generated, pressure-tested, and narrowed to the few that could actually survive contact with a market. By the final readout we had a room full of work that was, honestly, good. Rigorous. Defensible. The kind of thing you're proud to put on a screen.

And when it was over, the leader in the room looked at all of it and asked the only question that actually mattered:

"Okay. So what do we do?"

Not which slide was most interesting. Not whether the research held up. Just “so what do we do?”

I've thought about that moment a lot since. Because it wasn't a failure of the work. The work did its job. It was a signal about what leaders are actually buying when they buy strategy, and how badly the format we usually hand them misses it.

We've been trained to expect an answer

Instagram decides what you see next. Spotify queues what you hear next. Netflix tells you, with a confidence it has not earned, what you should watch tonight. These systems don't just reflect your preferences, but over time they shape them, and they train you to expect guidance that arrives without friction. You stop choosing from everything. You choose from what you were handed.

That expectation doesn't stay home when you close your personal entertainment apps and open your laptop at work. Customer expectations are transferable. The ease and responsiveness people get as consumers becomes the baseline they expect everywhere, service to service, product to product, industry to industry. And as the work gets more complex with more channels, more data, more decisions landing on fewer desks, the leaders I see increasingly want from their own organizations exactly what their personal algorithms give them as consumers:

Tell me what to do next.

That's where the idea of a Business Algorithm starts to matter.

What leaders are actually asking for

Underneath all the language of strategy, vision, and transformation, a lot of executives are quietly asking for something far more prescriptive. Not a 55-page deck. Not a quarterly offsite full of frameworks they'll never reopen. A clear, defensible recommendation that holds customers, markets, operations, and risk in the same frame and points somewhere.

They want decision support that behaves like the algorithms they live inside the rest of the day: context-aware, adaptive, always learning.

The catch is that business doesn't get the conditions those consumer algorithms enjoy. No clean data. No single narrow objective. No infinite room to experiment and let the model sort it out. A recommendation engine optimizes for one thing and gets a billion chances to be right. A leader is optimizing for growth, margin, brand, trust, culture, and regulation at once, and gets to be wrong roughly never.

So a Business Algorithm can't eliminate ambiguity. It has to work inside it.

It is not automation

Here's the distinction that everything else hangs on. The Business Algorithm is not built to replace judgment. It's built to accelerate it.

Consumer algorithms optimize for engagement, retention, time spent. A business algorithm has to optimize across constraints that actively fight each other. Which means the output can never be "do this, don't ask why." It has to be: here's what the signals suggest, here's why, here's the risk, and here's exactly where your judgment still has to do the work no system can do for you.

This is the part people miss in the rush to buy more AI, more data, more dashboards. The advantage was never going to come from having the most of those things. Plenty of organizations are drowning in all three and still can't answer a simple question on any given day. The advantage comes from designing a repeatable way to turn signal into action tha is fast enough to keep pace with customers, disciplined enough to avoid lurching from one shiny object to the next every quarter.

The work I described at the top wasn't missing intelligence. It was missing that last translation step. We'd built the map. The leader was asking for the route.

What a good Business Algorithm actually does

At its best, it helps a leader answer four questions, continuously, without having to convene a war room to do it.

1. What's changing? The weak signals and behavior shifts before they're obvious. 
2. Why it matters? The actual impact on customers, revenue, and risk.
3. What the real options are? Not infinite possibility, which is just paralysis with a nicer name, but a few curated, viable paths.
4. What to do next? With the confidence level attached, and the tradeoffs named out loud.

Notice that this is less about prediction than about prioritization. Less about certainty than about direction. It doesn't hand the leader a verdict and send them home. It sharpens the call they were always going to have to make themselves.

The Move

The organizations that win the next stretch won't be the ones with the most data or the most dashboards. They'll be the ones that built a repeatable way to answer the question their leaders are actually asking, the one many have been asking quietly for a while now and getting louder and louder:

What should we do next?