Becoming Data-Driven Is Tough—But It Doesn’t Have to Be

What does it mean to be a data-driven company? A data-driven company is one in which all decisions are based on knowledge obtained from different internal and external sources. In other words, no hunches are followed.

When you take a shower, you don’t get into the water until it’s the temperature you prefer. You might test it, cool it down, warm it up, test again and so on until it is the way you like. The concept is the same within a data-driven company. You check the info, make a change, check the info again until the results fit the expectations and improve the processes. Nevertheless, there are always changes. Maybe somebody turned on the dishwasher and the change in flow had an impact on the water temperature and you need to make changes in real-time to not burn yourself. Simple and easy, right?

For companies, things are more complicated. Each process is producing tons of data, either digital or analogical. When the indicators show something is not going well the Subject Matter Expert, or SME, takes action. Usually, the SME gets the solution right, but not because of the given data, but because of experience. The data is kept on the side because it was late, possibly incomplete, and in many cases, not trustworthy. Making the industry move on experience instead of facts. What would happen if your SME quits or a previously unknown problem shows up?

Similar to Dante’s Inferno, the industry moves within different levels of common misconceptions that prevent an industry to become truly data-driven.

Thinking “We are ready”

It is common to hear from management that they are ready. That the company bought the most expensive and futuristic ERP and MES systems, that the whole production is under control, and everything is working perfectly. Still, the industry is just a top performer in the era of mass production. Nothing is happening on the digital side. Are you able to see what is happening in real-time, take actions in real-time, predict possible disruptions, and reduce scrap, without having long meetings or groups of engineers trying to make sense of messy data sources?

“Nokia had an engine and infrastructure for innovating on the hardware front that was the pinnacle of the Age of Mass Production, but they did not have an effective engine and infrastructure for the Age of Software and did not have the management metrics or practices in place to realize that until it was too late.” -Project to Product, Mik Kersten

Feeling locked down

Again, it’s quite common to hear from companies that they don’t think they need a data-driven approach because they believe their current system is already giving them that. Once you get to understand the company more, many things come up. For example, we once worked with a customer that had a fully-fledged ERP across the company. Something really new and expensive. However, the data (while integrated) was not trustworthy. Information was misleading, timestamps were incorrect, a reconciliation took weeks, and tasks were handled by hand and a spreadsheet. Still, they wanted to continue with their provider.

Once they choose to go with their natural providers (ERPS and MESs) they hear “Of course you can do that’ ‘yes you can extract that data.’ But, it will cost you, you will need to buy a new module or build a new solution and then, before you notice, your budget is at the provider’s disposal and your needs are given priority based on your budget.

 Fear of taking action

How can you become a data-driven business if you don’t know where to start? It is simpler than you may think. You don’t need a large investment or a multi-year project. What you need is to take the first step and start investing in simple tasks that add value right away. For example, you can run a discovery workshop to understand the pain points of each area (quality assurance, production, management, etc.). Focus on the pain points that could have a simple data-driven solution, such as the production manager needing a solution that displays, in real-time, what is happening at the plant level. Work on defining a simple dashboard and deploy it. It may not be perfect, it may not be final, but it is the first step. You should take an Agile approach and find a trusted partner like Wizeline to guide you through the process.

In an Agile project, you will add value at every single iteration. This differs from a long-term project in which you can only see and evaluate after months of development. Becoming, on the long term, either not valuable or really expensive.

Fixing the wrong bottleneck

“Organizations that manage IT delivery as projects instead of products are using managerial principles from two ages ago and cannot expect those approaches to be adequate for succeeding in this one.”
-Project to Product, Mik Kersten

Many companies we have talked to suffer from the same problem,

  • Disconnection within teams
  • Software decisions and investments made without asking the SME
  • Consultants are only brought in when things are already going bad, really bad

If you start tackling the simple issues, without solving your real problem, you fall prey to small things that in general, never add up. Everybody will present small improvements in management meetings, everybody will applaud them, but everybody will also know that this is going nowhere. It may sound harsh but stay with me.

Data is your most valuable asset. Having it promptly, accurately, and orderly will allow you to make the right move—whether difficult or easy—and will make a positive impact. To have the right data, you will have to change things from the inside, automated recording, reduce errors and add an abstract layer that controls all that. The hard, but correct way, will lead you to a DevOps transformation that will serve as a foundation to become a data-driven industry.

Cost vs. investment

“Companies and their change agents still face big challenges in the pursuit of digital transformation, including a lack of digital talent and expertise (31.4%), the perception that digital transformation is a cost center and not an investment (31%), and general culture issues (31%).”
-Jim Hamill, 2017 State of Digital Transformation

In the digital era, digital solutions need to be seen as an investment. If your company leaders are not aware of the digital change and they are always pushing for cost reduction and seeing digital as a cost, then the company is doomed from the inside.

If your company has the vision and is willing to ride the wave, you need to work hand in hand to make the project a reality. Focus on your SME needs and increase ROI. It’s worth noting that not all IT areas are prepared to support a digital transformation. In reality, the majority may not even know how and where to start.
A technology partner is prepared to deliver MVPs in an Agile way that delivers value iteration by iteration, which can help you increase your ROI at a fraction of the cost.

Predicting the unpredictable

“With too little data, you won’t be able to make any conclusions that you trust. With loads of data you will find relationships that aren’t real… Big data isn’t about bits, it’s about talent.”
-Douglas Merrill, Former CIO and VP of Engineering at Google

Some of the objectives of a data-driven company are to be able to use its systems to forecast demand, availability, reduce downtime, improve logistics, schedule preventive maintenance; in summary, optimize your operation reducing cost. Artificial intelligence has become such a buzz word that it is easy to fall prey to incorrect implementations fed with incorrect assumptions.

In a couple of cases, we have observed the usage of ML algorithms trained with limited, unclean data, adjusted without the support of SME, and expected to solve impossible tasks. It’s no wonder why so many companies do not trust today’s ML-AI capabilities. The value of AI is achieved once you prepare the foundations to do it and put the time effort and correct team to implement it.

Our approach at Wizeline

The methodology we follow at Wizeline in which we connect different practices (Data, DevOps, Agile Development, and QA) allows us to act fast and simply. We have developed architectures that will simplify the process of connecting all your data sources, cleansing, stitching and presenting the data in a way that you can leverage it. We can help get your team ready and help fix what’s broken.

Long term, as you keep collecting your data, you will be able to start using advanced Machine Learning / AI solutions to make predictions, create models, simulations and optimizations, while focusing on your company’s pain points.

Data-driven industry, often called industry 4.0, is here but not all companies have been able to unlock the full benefits that come with it. Let us help connect your systems, make your data smarter, and allow you to start living the new industrial revolution.

Written by Luis De Alba, Head of Wizeline's Data & AI Practice
Written by Luis De Alba, Head of Wizeline's Data & AI Practice

Nellie Luna

Posted by Nellie Luna on September 17, 2019