Skip to main content


(This article originally appeared on LinkedIn Pulse, and can be found here.)

The world is changing. Since the digital transformation, everything is being quantified into data. This is probably the most profound technological shift of our lifetime because this data contains perfect information we can use to improve markets and continue to drive sustainable growth. To make use of this data requires a retraining of the workforce and a move away from a gut-only approach to a data+gut approach, where machine automation readily extracts the information you need to make better decisions. These are 8-steps you can take to consistently produce information to successfully act on.

# 1. Anoint a Directly Responsible Individual (DRI)

While everyone claims to be data-driven, most companies are regularly failing because data isn’t a true priority. Testing this theory is simple. If you don’t have a DRI for data in your organization empowered by clear data governance, you aren’t data-driven. This lack of priority also shows up in your accounting ratios. You probably know the percentage of revenue spent last year on marketing, sales, product, IT, cereal in the breakroom, etc. But what’s the ratio of data (including the DRI’s salary)? If you can’t answer, then chances are it’s close to 0%, meaning its practical impact is also close to 0.

#2. Audit Your Data Landscape

Next, is evaluating the current state of data in your organization. Odds are, things got pretty disorganized absent clear rules. Solve this by having your DRI conduct a data audit using an org chart, and visiting each group to identify stakeholders and their data. This should be a painless process involving one-on-one conversations instead of group interrogations, uncovering how stakeholders are currently using data, where the data comes from, and what they’d like to know (but currently don’t) from data. These conversations can be repeated for end users of data, like those depending on regular reports or dashboards to guide their work.

#3. Identify Your Data Maturity Stage

After having conversations with all the data stakeholders and end users in your organization, the DRI can identify your place in one of the four data maturity stages:

  • Human Reporting – this is anything produced by a human, usually using excel. To identify if this is your stage, ask what would happen to the next report if the stakeholder in charge of that data is hit by a truck? If there’d be no report then you’re at the human reporting stage.
  • Real-Time Intelligence – some information needs to be known in real-time, and some more impactful if reported automatically (real-time or not). This stage is easily identifiable by dashboards but often lacks human agents to investigate/change data consumed through visuals or reports.
  • Predictive Modeling – even if your data is flowing in real-time, it’s only telling you what’s happened to date. It’s completely possible to predict tomorrow with scary accuracy, assuming you have good data. This is usually done by data scientists and analysts that continually update their models in a manual fashion.
  • Machine Automation (ML/AI) – There are huge efficiency gains to using machines instead of humans for updating models. Machine learning does this through programmatic iterative processes that find, update, and deploy models to ensure the most predictive information possible. Done correctly, they require minimal human support. Ambitions of using AI are great, but realizing current limitations lets you create a plan and move up the maturity stages.

#4. Nail Your Data Governance to the Front Door

Laws are necessary for any functional system, especially one wanting to capture huge scales of economy via specialization. This is also true with data governance. Who gets to access what parts of the data? Who needs raw data to do analysis? Who just needs to see insights without doing anything else? What is data? Everyone in your organization should know the answers to questions like those above. Data is relatively new and evolves incredibly quickly, meaning adaptability is essential. If people keep breaking rules, chances are the rules need to be changed.

#5. Host SMART Goals Meetings

Have your DRI host a series of lunch meetings with decision makers, stakeholders and end users to identify goals for the newly identified and governed data. To ensure things stay civil, provide food and try the SMART goals system to focus the agenda of these meetings. Another method to increase success is the use of conceptual frameworks. This connects your data across departmental silos and explains it in relatable ways. The best of these frameworks are customer personas, the customer journey, and customer lifetime value, but there are others like the product lifecycle and the employee empowerment index. These force you to connect isolated data sets together in order to answer bigger organizational questions.

#6. Execute on a Roadmap

With this step, it’s time to start executing on data projects and moving up the data maturity stages. There can be some temptation to skip steps 1-5 and jump straight to this, but it’s a mistake responsible for more failed data projects than anything else. A plan, no matter how flawed, is still better than no plan at all. Steps 1-5 ensure the highest probability of succeeding given what’s within your control. This plan should start simple and be adaptive, but it should also include clear milestones with estimated completion dates (from the SMART goals). Then, it should be broken into tasks and assigned to specific individuals. Visibility of the roadmap is key to accountability, but you also need to show progress to keep people motivated to reach their goals.

#7. Sempre Avanti (Always Forward)

The reality is that no plan survives first contact with the real world. You won’t get it right the first time. Accepting this beforehand frees you from the burden of failure, while also making it possible to track the causes and adapt. Depending on why you fail, you may metaphorically have to take “one step back, for two steps forward” and revert to steps 1-7. Wherever you find yourself, simply repeat the necessary steps from there and try again. However, should you fail 3 or more times you have a serious problem and need to pause and reevaluate your DRI, governance, SMART goals, and roadmap because something isn’t working and it’s time to pivot your approach.

#8. Trust but Verify

To ensure you’re organized, your DRI should host regular evaluation/update meetings. Encourage openness at these meetings and provide an agenda covering what’s been done, current needs, and future actions. This keeps your team progressing in quick, iterative steps. The frequency of these meetings can vary, but ideally, these updates should become so streamlined that they turn into short meetings lasting 15 mins tops.

You’re Ready to Enter the Data Age!

If you find that your role is focused on “putting out fires” it’s a sign your work and quality of life will improve by handling your data through clearly assigned roles and responsibilities, data governance, a roadmap and regular meetings. But what is crystal clear from the history of digital transformation and data analytics is that 2018 is either the year you start moving forward in the data maturity stages or the year you officially get left behind.

Want to get the most out of your data? See our post on Interpreting Data Like A Pro.

Become a Data Demigod with our free lecture series