Stop Drowning in Discovery Calls: The Framework That Cuts Through 45 Minutes of Executive Chaos

How to turn a CFO’s data quality brain dump into one clear KPI and one actionable project

Last week, I did an intense data discovery session with a CFO that’s been dealing with data quality issues. Truth be told, this company is not ready for a data transformation (check out my article from December on knowing if a company is ready for a data warehouse or not).

Core sales processes are working, but manually tie back to Google Sheets. The workflows set up in their accounting and CRM systems are gapped and incorrect.

Information isn’t always entered completely in the source system itself, or else it’s entered inconsistently. The thread that ties everything together are these Google Sheets (manually hard coding values regarding customer information and billing). The company’s overreliance on these robust spreadsheets is its weak point. A result of scaling at lightspeed.

Stay High Level (Or Get Lost in the Weeds)

Anyways, I mention all this because I went into that conversation being told we’d discuss churn. However, I wanted to stay at a higher level to start. I’ve learned to never assume. I started asking the CFO about his objectives for 2026.

What resulted was a brain dump of information about all his data quality problems between the CRM and financial accounting software for 45 minutes straight.

Even 2 years ago, I’d have folded on this call.

The amount of side tangents that happen during these calls makes it extremely difficult to know what question to ask next. Not to mention, it dilutes focus on which problem to even solve. There are 1,000 problems.

One moment you’re trying to talk about 2026 objectives, the next minute you’re talking about the horrible data mapping process that sits in the Power BI semantic model which makes it difficult to accurately report who has churned.

Three minutes later the entire conversation flips from churn counts to a different metric. Maybe it’s Customer Acquisition Cost (CAC) or Customer Lifetime Value (CLTV), even just a basic Total Active Customers count.

I’d appreciate you to comment below if you have experienced this situation before. I think it’s a foundational experience for any data leader in the trenches.

So what do we do? I’ve put together a few steps combined with a little framework to help in these blurry situations.

The Discovery Framework: Your Flashlight in a Cave Full of Bats

1. Business Objective - What are you trying to achieve?

Like I said in the beginning, starting at a high level lets executive leadership open up from a visionary lens. It’s a low pressure way to get the conversation started without sounding robotic and remaining genuine.

2. KPI - How do we measure success against the objective?

This is critical for narrowing in on a specific problem. Why you might ask? Here’s what I’ve learned over the years. Executives don’t really know what they need. They’ll list off 1-3 objectives and then 50 KPIs to track.

The key lesson…

Always bring them back to one KPI at a time. If an executive’s blabber is a cave filled with bats and monsters, narrowing in on one KPI at a time is the light at the end of the cave. It guides you.

3. Data Needed - What data do we currently have?

Which source systems pertain to this objective and KPI? Will we have to combine data together to see the full picture? Start thinking about that architecturally.

4. Current Pain/Gap - What’s broken that prevents you from achieving this?

For more mature organizations that have a digital process, it might be a slight tweak to their Hubspot and TypeForm process.

For less mature organizations, it might be an entire overhaul of their sales process and system workflows (like in the case of my experience last week).

If it’s an entire overhaul, question if they’re even ready for data solutions at this point of their journey. Don’t be afraid to explain they’re not ready and walk away.

5. Step Forward - What’s the actual fix?

Garbage in garbage out. If you’re gauging that it’s completely broken at the source, offer to work with them to rearchitect those processes completely.

Align these rearchitected processes to a Business Data Model. That offers clarity and alignment on data definitions and process structure without the technical complexities of building a logical and physical data model right away. They’re not ready for that yet.

You can explain to them that it’s critical for a data professional to help with these foundational pieces from the start so by the time the data is generated properly, everyone has full context and alignment.

This is about bridging that gap from being an IT data engineer to being a strategic partner.

If it’s not totally broken at the source, it could be a nice opportunity to build a data solution that helps them track CAC and how to keep costs down. Maybe it just demands a few tweaks to their current TypeForm fields to track customer acquisition more accurately.

With either of these, go back to your data lab and map out a timeline of steps for implementation and send it over to them.

---

The Secret Weapon: Come With an Agenda

Shout out to a recent article by Madison Schott. She pointed out the importance of coming to data meetings with an agenda, and it sparked a reminder at just how critical this is. Not only does it help guide the conversation, but it helps you be prepared in front of highly time-sensitive executives. Trust me, they appreciate this.

Here’s an example agenda:

I. Goals/Objectives for 2026

• Why Churn and CAC as priority?

• What decisions do these drive?

II. Metric Definitions

• Churn definition

• CAC definition

• How do you want these reported? (i.e. grouped by customer cohort, time/date, etc.)

III. How are these metrics currently tracked?

• Source(s)

• Who owns the process today?

IV. What are the difficulties/limitations of the process?

• What questions can’t you answer today?

V. Are the current state metrics trusted?

• How are metrics reconciled if different?

VI. Scope & Path Forward

• Deliverable

• Next steps

---

The Uncomfortable Truth

Hopefully this helps in your journey honing quality data discovery that helps you lead the discussion! None of these things feel natural initially. It will be uncomfortable, it will feel ambiguous, and you’ll make mistakes.

Just know that focusing on a single KPI as your north star often helps eliminate the noise, allowing focus and clarity in what the real issue and potential solution is.

---

Christian Steinert is the founder of Steinert Analytics, helping healthcare organizations turn data into actionable insights. Subscribe to Rooftop Insights for weekly perspectives on analytics and business intelligence in these industries.

Feel free to book a call with us here or reach out to Christian on LinkedIn. Thank you!

Also - check out our free Healthcare Analytics Playbook email course here.