As a CEO, it is more than likely you are under increasing pressure to demonstrate a tangible return on your analytics strategy. But for many CEOs, the ROI on data has been elusive.
Data scientists are hired, resources invested in analytics software, and data warehouses built and rebuilt again. And even though your leadership team can't quantify the bottom-line on the last set of investments - they urgently need budget for a new round.
Budget for Tableau. For AWS. For a data lake. For machine learning training.
It's frustrating and uncertain. But here's the thing:
Some of these challenges actually start at the top, and as CEO you have the power to fix them.
Here's a quick look at 7 troubling indicators that may indicate you're part of the problem.
#1 Business Model
Is your business model clearly defined? Does your leadership and your rank-and-file have a shared understanding of how you create value as an organization - and where to optimize and innovate? Or are they working on what they think is important?
If they aren't focused on your North Star, the data projects may be interesting - but they won't be pertinent and the organization as a whole is unlikely to implement them or realize true value.
If your business model remains in flux or is poorly understood by your IT team, it is difficult for them to measure the business model's vital signs - let alone create projects to optimize it.
In your next conversation with a data scientist or an IT Director, ask them to whiteboard how the organization generates value and what the friction points are.
#2 Too Many KPIs
When you count everything, you count nothing. If you have more than 3 KPIs for each major performance question (KPQ), you are probably in danger of diluted focus.
A simple test for too many KPIs is if you can even remember what your KPIs are. Right now, could you write them from memory on a whiteboard?
And can you articulate what business model questions drive them? Or are they a snowball collection of KPIs by committee?
Working through the rationale of the KPI structure - and tightly coupling them with the business model will focus your team and ensure they aren't working at cross purposes. And it might be scary, but it is not unreasonable to see if your leaders could also whiteboard them from memory.
The company KPIs should be "in everybody's bones."
#3 KPIs Overly Oriented to Financials
It's a problem if your key performance indicators relate primarily to financial results.
Your business model revolves around the profitable delivery of value, through whatever special sauce your organization provides. Value should be where your KPIs focus because that's where the seeds of your future financials are planted.
Most financials tell us about the past - and, yes, they can point to satisfying or alarming situations. What they don't do is tell you how well you are delivering value in your key relationships: with your clients, employees, channel partners and vendors.
Your business model is its own particular organism with its own unique vital signs. When these vital signs are healthy, your company is healthy. Financial KPIs sound an important alarm, but they don't replace quantitative insight into the challenges of delivering future value efficiently.
#4 Siloed Data Teams
Watch out if different departments are creating redundant solutions, data warehouses, and data science teams. If there is infighting between different groups around who should be building what, then you are probably spending too much money in too many places. It is also likely they are operating at cross purposes, ignoring the value in each other's data stores, and reinventing the wheel.
When costly data projects are managed through an enterprise data portfolio, redundancy and waste are reduced, key resources are aligned with organizational needs, and your likelihood of maximizing your most expensive resources on the most critical problems is increased.
If you haven't created a prioritized enterprise data portfolio to manage your different data initiatives, there's a strong chance you're at risk. You probably have redundant data sources, top resources working the wrong priorities, and departmental teams functioning in silos.
#5 Hazy Outcomes
Some quick level-set questions:
- Can you stack rank the ROI on the last year of data initiatives?
- Do you deliberately abandon and unwind unprofitable data initiatives?
- Were your data initiatives framed with a target ROI? Are these promises captured?
- Are the business and technology teams that deliver data initiatives accountable for results?
- Are your data initiatives tied to measurable business targets?
As the primary investor in your company's analytics effort, you have a reasonable expectation to know which projects are delivering value and which aren't. You should know which stakeholders deliver value and which don't.
You deserve a clear bottom-line on the ROI of your data investments - and as CEO you help your team by setting this expectation before investing. If you're not sure if you're getting value from your data and can't quickly summarize how much, you're operating blind and may be in danger.
#6 Freeform Analytics Initiatives
Data scientists love patterns and working with data. When they are good, they can tease out amazing correlations and insights. Presentations of what they've learned can be riveting.
What they aren't, generally speaking, are business people. They aren't focused on the efficiency and value your business generates. They are focused on insights, patterns, and the stories that can be generated from your enterprise data.
This fascination with interesting patterns data might be a risk for you.
If what they are working on is "off task" from your strategic priorities, you aren't running a business, you're running a research facility. And your business community is highly unlikely to operationalize their insights when they don't address the targets they are incentivized to deliver.
What is interesting doesn't get done. What is measured gets done. You need alignment throughout the entire value chain of working with information - which isn't emphasized enough.
So here's a question for you:
When you went to the last data presentation in your organization, was there a clear relationship between the work and specific business goals?
#7 Magical ThinkingMagical thinking - which is everywhere in organizations these days - sounds like:
- If we buy Tableau/Domo/PowerBI, we'll get a data culture and everyone will deliver value
- If we build a data lake and fill it with all the data we can get our hands on, value will come
- If we build a data warehouse, the business will be able to ask any question it wants
- We urgently need data scientists or AWS or blockchain or Watson or AI/ML/DL to help us catch up
- I'm not sure yet what I would put on it, but I really want a CEO dashboard
- Somebody or something else will deliver the ROI we've all been waiting for
CEOs create a data culture that treats data projects as business investments. They insist on business plans for investing in data initiatives. And they spread the word that the creation of value through data is a thoughtful and deliberate business endeavor.
How Does My Role as CEO Fit Into The Strategic Analytics Lifecycle?
It's reasonable to wonder how the CEO's role participates in the overall value chain of data ROI. As a quick referennce, you'll find a visual overview below.
If you're interested in learning more about the Bartlett System Strategic Analytics model, we have a downloadable "CEO Playbook" slide deck to help frame the scope of your responsibilities - and those of your team.
And feel free to reach out to me personally if you're interested in discussing your strategic analytics challenges.