How the French Army of World War 1 Can Teach You to Run a Better Analytics Team

100 years ago, this month, WW1 ended. A bloody four-year long war that changed the way we live. As is with many great struggles throughout history, innovation is something that tends to grow quite rapidly and WW1 was no exception to this. As a result, many new ideas were put into practice as each competitor tried to outdo the other. Tanks were one such invention that grew from the great war. Most often we think of the British and their tanks or the massive German land ships. However, it was the French who we can credit with some of the innovations that we still live with today. And it was those innovations that can be lessons to analytics, big data and data science teams today.

In 1915, the French went about building their own tanks as they thought it would be a great way to break the stalemate on the western front. Their first go at making tanks was very expensive with the first model expected to cost 7 million francs for 100 tanks, which was a lot of money back in those days. Those first tanks had to compete for resources. French industry was maxed out producing other arms and steel plates had to be imported to build the tanks. Eventually a 60-horsepower tank was created that could reach 7 kilometers on level smooth ground and half that on rough ground. The armor was just thick enough to block small arms fire but not a hit by cannon fire. These tanks were also death traps as they were not designed for safety of the crew with the engine being places in front, which is where most of the impacts of hits were taken. In other words, these tanks were not perfect. Kind of a lot like the first big data and data science teams.

Those early teams were fun, I enjoyed the time I had, yet what we were creating had a lot of improvements to be made. We had rough ideas and idealism, yet the ability to make data science and big data work for an organization as a business unit was still a way out. Many people believed that these new disciplines would replace everything else and sometimes when I visit a company, they still think these thoughts. But we know that isn’t true and our rough first go at these new analytic fields needed improvements. Companies were disappointed with results; team leaders were frustrated with the lack of resources and team players were often out of touch with what was really expected. Something new and different was needed.

In the spring of 1916, the French realized that their first go of tanks needed improvements and someone came up with the bright idea of, lets just make it bigger! A crew of 9 for the new tank was envisioned and bigger guns, more guns! Hell, even a flame thrower was dreamed up of being put on this tank. It would eventually be called the Saint-Chaumond tank, if you played Battlefield 1, you may be familiar with it. It was bigger and heavier, but its tracks were the same size as the first tank created the year before. And as anyone knows, that means more weight on the tracks which means it can sink in the mud a lot easier.

As analytics grew, over the past few years, we saw this happen with data science teams and big data teams as well. They saw their failures as a way to grow. We need more people, faster systems, bigger systems! We need a seat at the executive table. The importance of the chief data officer or the chief analytics officer grew. But failure was still the norm, mainly because like the French’s new tank in 1916, the foundation of how to make analytics work, hadn’t changed with the times. In 2016 to present, companies were seeing some minor success but nothing that was really grand in scale. Sure, there are small wins here and there but the ability to scale it out and use data in ways that create rapid scaling was still a dream for many. The French in 1916 introduced tanks on the Neville offensive and saw massive failure as a whole. Yet they saw individual tanks perform great and used that knowledge to change their tank strategy and methods of usage.  Tanks were used as shock weapons to roll up with infantry and help clear out enemy soldiers. This didn’t work, too many tanks were getting destroyed and soldiers getting killed.

The French went back the drawing board and realized that the main killer of tanks was artillery. So that meant that taking out artillery needed to be the main priority. Since destroy artillery is the main priority, that means you need to know where they are. That means you need to have aerial reconnaissance in the form of aerial photography. In order to make aerial photography possible, you needed to have local air superiority. Once they had air superiority, they needed a high concentration of their own artillery to neutralized enemy artillery. Once the artillery takes place, the tanks need to know their routes over the best possible ground, in advanced, with smoke shells to help them get as close as possible to the enemy positions to help minimize artillery taking out too many tanks. They needed better communication with infantry which meant new communication methods with other parts of the army. And finally, tanks would be used with limit engagements and limit goals.

When the French used their new approach to engagements, they were very successful in reaching their goals. In their first engagement, many tanks broke down, which was common back then. But the tanks played a vital role in breaking lines or drawing attention away from the infantry, allowing the infantry to reach their objectives. 

Like the French, many companies need to think about how they use data and really move away from a siloed approach to their teams. Often teams are created as a new silo in the company with little to do with everyone else in the company. The French realized that was a disaster and they changed their approach to have a more integrated approach. Most data teams need to do the same and spend time communicating with the various touchpoints of their organization. Because like the French learned 100 years ago, you can have new tech and great tech but if you can’t communicate with the rest of your organization, that tech is a waste.

But the French didn’t stop there. They had heavy tanks and they decided they needed light tanks. Tanks that had a crew of two, a rotating turret, which was new at the time. This tank was also much lighter, only 4 tons and could travel twice as fast as any other thank at the time. Add to that, the tank was easy to produce. The FT’s were what most people called these tanks. They were easily deployed from June of 1918 and the French produced over 2,000 of these tanks, more than the other tanks combined. In fact, the FT is often called the tank that won the war. The French combined forces doctrine didn’t change with the FT, in fact they were able to take more ground faster with these tanks than the enemy could possible hope to recover as the speed of the tanks make it difficult for regrouping to take place for a possible counter attack. The FT design is also what is often called the template for all tanks even up to this day. The basic design of two tracks, a moving turret with a gun, is essentially the same 100 years later. 

Data teams needs to find their FT. Right now, we don’t really have one because most companies haven’t had the conversation about their combined forces strategy to create the FT scenario, that will give them the power to compete against their competitors. As a result, data science and big data is still a trench warfare style endeavor. 100 years after the great war, we can learn lessons from the experiences of people who lived back then and apply them to current situations. The French evolution of tank use is one that can teach us the importance of blending our teams and the importance of innovating for real problems, not just for the sake of innovation in itself. Something data teams need to learn today if they hope to survive. Because trench warfare was never good on the soldiers in the trenches or their respective countries.

I want to thank the youtube channel, The Great War, which is where I got the research information about the French army. I highly recommend checking out their channel.