Translating Data into Better Decision-Making: Customers and Innovation

Most people these days associate innovation with technology. Yet in a lot of cases, it is not the technology that really makes the difference: it is the deep understanding of customer needs. D’Acunto explains: “Sure, technology can be a strong enabler of novelty, but it’s not the only factor for success. Organizations need to evaluate the needs of the customer and the ways they are not being met yet. Only then can they find innovative ways to solve these issues, making use of technology if it fits their solution. In other words: companies often have to take a step back and look at issues that might have been there the whole time, before the technology to solve them had even been invented.”

A practical approach
So, which practical approach should you take? D’Acunto thinks every company should start by doing market research. “This can highlight what’s going well and what should improve, and delivers precious knowledge about the needs of the customer,” she adds. “Often, available market research provides some good pointers. In addition, interviews with customers are a powerful instrument to help you understand what your customers are trying to achieve in a broader sense – customer journeys and jobs-to-be-done are for instance ways to look at the full spectrum of the customer wants – which in turn helps you to uncover what triggers them. On top of that, companies should assess which problems are worth solving, in order to make a difference for the customer. Only then can you start thinking about a solution.”

Sometimes, D’Acunto thinks, companies don’t focus enough on answering the questions that really matter. “What is the problem you are trying to solve? Companies can lose sight of what is important. Sometimes, their appetite for innovation is driven by the urge to be able to say that they are using machine learning and AI, for example. Instead, it should be driven by a clear framing of real problems and real needs. You shouldn’t use a chatbot just because your competitors are doing so.”

To find out which problems are worth solving, you not only need to understand the customer. You also need an understanding of (the future of) the business. “That is why we approach innovation projects in collaboration with the customer,” says D’Acunto. “They know where their business is going, we know where data science is going. As a data scientist, you have a good idea of the methods that are out there. You know which technologies are suitable to fix a given problem. The real power of innovation can be unlocked by working together and combining each other’s strengths. That is why it is important to have data scientists on board early, when you’re still generating ideas.”

_________________________________________________________________________________________________

“The most important thing we learn from design thinking is to keep the customer central in anything that we design.”

_________________________________________________________________________________________________

Always keep an eye on the customer
A study by SHRM's Executive Network shows that out of a thousand ideas, only one will typically survive in three years’ time. “That’s why it’s very important to generate as many ideas as possible in the brainstorm phase,” explains D’Acunto. The most important thing we learn from design thinking is to keep the customer central in anything that we design. This focus on the customer helps to identify the best ideas, and it remains relevant throughout all phases of innovation. Even if it is something that just optimizes processes, it can always be translated into something that has an effect on end customers, like faster deliveries or lower prices. Regarding AI and innovation, there are three things a company wants to learn:

1. Desirability
   What the customer wants
2. Feasibility
   Whether it’s possible given the technology, legislation, data, et cetera
3. Viability
   Whether it contributes enough to your business model

These learnings can be approached one by one.

Low-tech beginning
Companies often think that limits will be set by technology. They have ideas and want to know if these are feasible. D’Acunto has reservations about that approach. In her experience, there are hardly any limits to what technology can do. But you have to ask yourself these questions:

• Is it really what the customer wants?
• Is it what’s going to make the difference out there?
• What kind of business model are you going to use?
• Will it be accepted by society?
• What is the effect of regulation?

“All these factors might impact the success of the project,” she adds. “Feasibility is not the only thing you need to think of. A project should also be desirable and viable.” Desirability and viability are not things you can prove in a lab, which is why D’Acunto favors a modest beginning. “Before you invest in technologies, you want to understand what the customer wants and validate that. You can do so by offering a simplified, low-tech version of the product to a restricted group of clients and measure their reactions (desirability). Then, you can find out how well you can achieve this goal with the high-tech version (feasibility) by offering it to the client and seeing if they are willing to pay for it, or if it indeed generates the kind of outcomes you expect: a sale, or a certain behavior (viability).”

_________________________________________________________________________________________________

“The whole point is to identify the riskiest assumptions in your solution and validate them as quickly as possible.”

_________________________________________________________________________________________________

When experimenting with AI, we are talking about end-to-end experiments. You have to be able to reach the client. D’Acunto explains: “The whole point is to identify the riskiest assumptions in your solution and validate them as quickly as possible, in the simplest version possible. If the outcome is positive, then you can take that feature and integrate it. In addition to experimenting on a product level, you can also do so on a feature level. If you have an app for customers, for instance, and want to know if a feature is a good idea, a low-tech solution initially works best.”

_________________________________________________________________________________________________

“Companies accelerate changes to their business model that would have happened anyway in a year or two.”

_________________________________________________________________________________________________