Artificial intelligence is dominating nearly every conversation in business today.
Depending on who you ask, AI is either going to revolutionize every industry or it’s an overhyped tool that’s little more than a faster search engine.
The truth, as is often the case, is somewhere in between.
One of the more interesting discussions I’ve seen recently centered around a simple idea: AI can provide information, but it can’t make decisions for you. I mostly agree.
But I also think we’re asking the wrong question. Instead of asking whether AI can replace decision-makers, we should be asking: How can AI help people make better decisions?
That subtle difference changes everything.
We’ve Never Had a Data Problem
I’ve spent much of my career working with business intelligence, analytics, and travel management technology.
Long before generative AI arrived, organizations were investing millions of dollars in collecting and organizing data.
We built data warehouses.
We created dashboards.
We normalized information from dozens of sources.
We defined KPIs.
We tracked trends.
We measured everything we could. And yet…
Organizations still struggled to make timely, confident decisions. Why? Because data has never been the bottleneck. Understanding it has.
Dashboards Tell You What Happened
Traditional business intelligence is incredibly valuable.
A dashboard can tell you:
- Airfare increased 8%.
- Hotel compliance dropped.
- Average trip cost is higher than last quarter.
- Online booking adoption improved.
That’s useful information. But information alone rarely answers the question executives actually care about: What should we do next?
That’s where AI begins changing the conversation.
The Five Layers of Decision Making
I like to think about business intelligence in five layers.
1. Data
Raw facts.
Hotel rate: $289.
2. Information
Context.
Hotel rates increased 8% compared to last year.
3. Insight
Understanding.
Most of the increase came from travelers booking inside five days.
4. Recommendation
Suggested action.
Increasing advance purchase from five to fourteen days could reduce annual hotel spend by $400,000.
5. Decision
Human judgment.
Should we actually change our travel policy?
This final step is where experience, leadership, and organizational priorities still matter.
Travel Management Is a Great Example
Travel programs generate an incredible amount of data. Every reservation tells a story.
- Where employees travel.
- How far in advance they book.
- Whether they comply with policy.
- Which suppliers they choose.
- How disruptions affect costs.
For years, travel managers have relied on dashboards to monitor these trends. Imagine what happens when AI joins the process. Instead of simply reporting that hotel costs increased, an AI assistant could identify the underlying reasons.
- Perhaps a major conference drove demand in one city.
- Maybe negotiated hotel rates are no longer competitive.
- Maybe travelers are consistently booking at the last minute.
Rather than stopping there, AI could recommend several options.
- Increase advance purchase requirements.
- Recommend alternative preferred hotels.
- Adjust nightly rate caps for specific cities.
- Renegotiate supplier agreements.
- Monitor the impact over the next thirty days.
That’s a very different experience than staring at charts and trying to connect the dots manually.
The Future Isn’t Better Dashboards
For years, we’ve focused on making dashboards prettier.
More colors.
More charts.
More KPIs.
More filters.
But executives rarely wake up wanting another chart.
They want confidence. They want clarity. They want to know: “Given everything happening today, what’s the best course of action?”
That’s where AI has enormous potential. Not because it replaces analytics. Because it builds upon analytics.
Dynamic Travel Policy
One area that particularly excites me is the future of travel policy. Today, many organizations review their policies once or twice a year. The business changes much faster than that. Imagine instead a policy that’s continuously evaluated.
AI monitors:
- airfare trends
- hotel pricing
- supplier performance
- traveler behavior
- compliance rates
- disruption risks
- sustainability goals
When something changes, AI doesn’t immediately rewrite the rules. Instead, it recommends adjustments. Perhaps hotel caps should increase temporarily in London because of a major international event. Maybe advance purchase should be tightened in Chicago after several months of rising costs. Perhaps a preferred airline no longer offers the best value.
The travel manager reviews the recommendation. Approves it. The updated policy is automatically published to the booking platform. Employees immediately benefit from smarter guidance without waiting months for a policy review cycle.
That’s not science fiction. The technology is rapidly moving in that direction.
AI Still Doesn’t Understand Everything
As exciting as this future sounds, there are decisions AI shouldn’t make alone.
AI doesn’t fully understand:
- organizational culture
- employee morale
- executive priorities
- customer relationships
- political realities
- long-term strategy
Imagine AI recommending tighter travel restrictions that save money but make it harder for sales teams to build client relationships.
On paper, the recommendation looks perfect. In practice, it could hurt the business.
That’s where human judgment remains essential. Leadership is rarely about finding the mathematically optimal answer. It’s about balancing competing priorities.
The Best Decision Systems Will Combine Both
I don’t believe AI will replace travel managers. Or analysts. Or executives. I think it will make all of them better.
The organizations that gain the greatest advantage won’t simply have the most data. They’ll have the best decision systems. Systems that combine clean, trustworthy data…Powerful AI…
And experienced people who understand the bigger picture. That’s a much more compelling future than either humans or AI working alone.
Final Thoughts
For decades, we’ve measured the success of business intelligence by the quality of our reports. I think the next decade will measure success differently. Not by the number of dashboards we build. Not by how many KPIs we track. But by how effectively we turn information into action.
AI won’t eliminate the need for judgment. If anything, it makes good judgment even more valuable. Because the organizations that succeed won’t be the ones with the smartest AI. They’ll be the ones that know when to trust it, when to challenge it, and when to make the final decision themselves.
The future of business intelligence isn’t about replacing people.
It’s about helping people make better decisions.
