How to Think About Data Value
Roughly a week ago, I was presenting a data warehouse evaluation and implementation plan to a client. The meeting was set up as a high level technical review of their current warehouse needs and our proposed implementation plan for approval.
I presented the solution that the Lakefront team had prepped and outlined the major components of the ingestion, transformation, storage, and reporting aspects of the implementation to the clients executive team.
For the first half of the meeting the client seemed engaged and asked pointed questions about the choice in ELT tool and warehouse. As the call went on I could tell that the CEO’s patience was running short. We had gone down two or three rabbit holes about the merits of AWS vs Google managed warehousing solutions and their relevant cost structures, and he seemed to be getting squirmy, likely thinking about his next meeting or task to complete for the day.
In the final minutes I found us behind schedule and had to rush through the last slides in the presentation. Once I finished the presentation, the CEO turned to me and said, “So we are going to talk about what we are doing with all this new data at some point right?”. A little taken aback by the question, I thought to myself how we discussed that this meeting was the technical review and how we had already had discussions about the ways we planned to use the newly available data to create unified customer, lead, and opportunity datasets that would allow for more robust near real-time reporting on their business. We had also discussed how once we had these datasets, we could produce machine learning models that would help them sort their leads by likelihood to convert and group their customers by purchasing behavior.
I could tell that, while I had a clear vision for these use cases, I had not done a great job reiterating them to the clients senior management team and more importantly, giving them a framework for measuring the impact that these new capabilities would have on improving their business processes. Expectations had been set for the project milestones but an outline of the value that they were receiving for this work had not been solidified and made available to them.
Missing the Value
While one could make the argument that a business should know what they want from their data before engaging a partner like Lakefront to implement a warehouse or other service, the reality is that it is not that simple.
Oftentimes business leaders can get “lost in the sauce” following industry hype and feeling the FOMO at the country club when their buddies tout about their new AI model that's taking their business to the next level. Other times they may have a good idea of how data may be able to help them with their business but can only conceptualize their thoughts at a high level, and may need a partner like Lakefront to fill in the details.
Either way, there are many consultants / agencies that will sell the data dream but not have a framework to measure the actual impact that their work has on the clients business. Worse actors may actively avoid the conversation and prey on a client's lack of understanding of how to measure Data Value and deliver mediocre results on a continuous basis, banking on the fact that no one at the clients organization will be able to challenge their work with a measurable goal to set them against.
What is Data Value?
Data Value, like any great concept, is simple. It boils down to one question, what is the measurable improvement that data resources at your organization have on the business? If you ask any leader of a business there are only two core concepts that qualify as improvement, increase revenue/profits and decrease costs. All other concepts of improvement fall under one or both of these buckets.
Starting at such a base level may seem obvious but many leaders struggle to root projects or efforts that their internal teams are making to one of these concepts, and then be able to use those core concepts to measure the success of their efforts.
To understand how one does this with data we can exercise an example below:
Use Case:
A business wants Lakefront to create a data warehouse to create automated reports measuring sales team performance. This is currently done through a manual report that the sales director puts together each month which takes her 10 hours to complete.
Data Value:
Increased Revenue/Profits - An automated report that updates daily created by the Lakefront team would allow the sales director and her managers the ability to review daily sales results with their teams. This change from a monthly review should allow the team to address issues as they arise instead of long after the fact, and increase the teams performance converting sales on a continuous basis.
Decrease Costs - Creating an automated report that is continuously feeding up to the day data vs a manually created report saves 10 hours a month for the sales director, if this person was paid a salary of $150,000 this would equate to $800 / Month in saved time, more than paying the cost of maintaining a modest warehouse solution. Not to mention the fact that the director now has those 10 hours back at the end of the month that they can use for other more important tasks.
VALUE MEASUREMENT:
Measuring the value of this implementation can be done in several ways. The first is by employing the time savings calculation listed previously.
$800 x 12 Months = $9,600 Annual Cost Savings
The sales team performance would be a more difficult measurement to perform, this is mostly due to the fact that changing sales performance through the use of the automated report requires the business to actually buy in to using the report on a daily or at minimum weekly basis. Also there are external factors that influence sales team performance like market factors.
With that said, Lakefront would still report on sales performance metrics before and after the sales report implementation, likely with a combination of anecdotal feedback from the director and management on how the new report has influenced their weekly operations. For this example, say the company sold 5000 widgets each month for $50 a piece before the report implementation and after they sold an average of 6000 widgets.
1000 increased widgets x $50 x 12 Months = $600,000 Annual Increased Revenue
Total Data Value = $609,600 Annually
If the total cost of the project to set up the data infrastructure and automated report was $100,000 for the implementation and $560 / Month to maintain the solution, the company is making an estimated $502,880 ($609,600 - $106,720) annually after all is said and done.
Conclusion
It isn't a perfect science but setting a measurement plan in place and following through with reviewing the results of that plan holds both an implementation partner and internal stakeholders accountable for the success of a data focused initiative.
After completing our review last week I followed up with the senior leadership team and let them know that we will be setting aside time to flesh out our use cases for their newly implemented data warehouse focusing on the Data Value they should be expecting once all is said and done.
If you're unsure how your company is measuring the value of their data focused efforts Lakefront can assist with putting together a framework and following through with measuring your Data Value, reach out to see how we can help.
