Contact Us
Contact Us

Expert view

4 steps for leveraging data to solve aviation problems

January 22, 2020

This article is an excerpt from a Cirium on-demand webinar, Moving your strategy beyond the walled garden: How airlines and […]

This article is an excerpt from a Cirium on-demand webinar, Moving your strategy beyond the walled garden: How airlines and their service partners can take control of their 2020 results. You can download the on-demand webinar here.

I want to tell you a story from a different industry that I think can serve as an analogy and will help you get ready to deal with these new problems the commercial aviation industry is facing.

A number of years ago, I was called into help a large law enforcement agency do criminal gang analysis. They wanted to better understand who was in specific gangs and how they were connected.

The team at the law enforcement agency were used to using certain kinds of data. In law enforcement and national security circles, like in commercial aviation, they have data that they’re comfortable with and consistently use in their problem solving.

But there’s an old saying that goes: The best place to solve a problem is not where the light is best.

If the light is shining bright somewhere, then it is unlikely there are a lot of problems left to solve because the light is already there. You want to go off into some place less comfortable where there is less obvious data to help see your problem. That was the problem the law enforcement agency was facing. They didn’t have a way to link these entities together to figure out who was connected and how.

Step 1: Identify the actionable problems 

We started with their shared problem because this was a project that had a consortium of dozens of different law enforcement agencies involved.

I asked them: “What is your ultimate goal?

Their responses started with statements like, “Well, we really want to stop human trafficking through these gangs.”

I had to stop them right there. It was my job to be the bearer of bad news and tell them that’s not an actionable goal. When designing an actionable goal, you need to start by thinking about what it is you need to do. Then, consider what individual actions you need to take in order to achieve that goal.

Eventually, we got down to the goal of arresting more people to seize more assets and do it faster. Those are things you can act on. That’s what I mean about shared problems that are actionable.

Step 2: Find related data sets

To address their actionable goal, we started to look at the data they had. It was clear they didn’t have the data they needed. You need to figure out who’s connected so you can roll them up in one big sweep. You need to figure out what they own and as a result you’re not going to be able to go faster.

I guided them through the process of finding additional data sets completely outside what they normally dealt with. We collected data from sources such as property records, credit records, etc. Because what they were used to was essentially a combination of what was in their arrest record systems and phone records. And that’s just not what they were trying to accomplish. They didn’t care who was calling whom. They wanted to arrest people and take their stuff as a result.

And that’s what helped them. Together, we built a system designed on new data because we were able to define a whole new problem.

Step 3: Be as specific as possible

The moral of this story for the commercial aviation industry is you need to start with the person and get up to the network. Commercial aviation has already pretty much nailed how to start from the network and now it needs to go down to people. It’s almost a mirror image of the exact same problem. Two sides of the same coin.

What I would encourage people to do if you’re trying to be the first to spot opportunities is to be very clear about exactly what kinds of opportunities you’re trying to spot.

If you say, “I want a better share in a particular market.” Honestly, I’m not sure that’s good enough because all of your competitors are doing the same thing. You need to get more specific.

Do you want to sell more tickets to people who will buy those tickets at the front of the plane? If that’s the case, then you need to be focusing on those kinds of people and you need to know more about them.

Do you want to make information more accessible so that your revenue management team can price more specifically for those same people?  Well, if that’s the case, then get very specific around what kinds of price actions you want those people to take.

Step 4: Focus on the individual, active entity

Commercial aviation has done an amazing job of pioneering network analysis in business. What it now needs to do is go from seeing things as only a network, to seeing thing in terms of the individual consumer — see that network as simply a way in which people decide to connect with other people. As a result, you’ll be seeing an old problem in a new way.

The old problem is you want more people flying on your planes. The operative word there is people. I would encourage you to think of every single consumer sitting on your plane as an individual with specific needs, preferences and behaviors. Stop thinking of them as passengers, because a passenger is a passive thing. A consumer is an active entity. They make choices and you can see what kinds of choices consumers make.

If you’re willing to see the old problems in a new way and are comfortable bringing in new kinds of data, then you’ll start solving more problems and you’ll solve them faster than the competition.


Interested in what the Cirium data portfolios and smart analytics can do for your business? You can find more content here. Read about our innovations and solutions to develop commercial strategy and improve the traveler experience.

RELX logo