Stakeholders Aren't Happy — Even Though Data Teams Are Working Hard

In most organizations I work with, the problem isn't effort.

Data teams are busy.
Tickets are moving.
Dashboards are shipping.

And yet—stakeholders are frustrated.

They don't trust the numbers.
Insights take too long.
Every meeting turns into a debate about whose metric is "right."

This is one of the most common patterns I see when organizations inherit or attempt to modernize a legacy data platform.

The Real Problem (It's Not Talent or Tools)

When you look closely, a few root causes show up again and again:

• Data is siloed across teams, systems, and tools
• Metrics aren't trusted, because definitions vary by report or team
• Time to insight is slow, even for simple questions
• Stakeholders feel blocked, despite heavy data team effort

Behind the scenes, data teams are usually stuck in a reactive loop:

One-off requests
Custom reports
Manual fixes
Firefighting broken pipelines

Demand keeps growing, but there's no time to build leverage. Engineers stay busy, but progress feels flat.

This is the moment where leadership starts asking: "Why isn't all this investment paying off?"

Changing the Narrative: From Reactive Support to Proactive Platform

The turning point comes when teams stop treating data as a support function and start treating it as a product platform.

In successful transformations, I've seen three deliberate shifts make the difference.

1. Self-Service by Design (Not by Hope)

Instead of every question flowing through engineering:

• Standardize ingestion and transformations
• Define metrics once, centrally
• Publish governed, well-modeled data products

When business users can answer most questions themselves, engineering regains focus—and trust improves organically.

2. Conversational Analytics to Kill Ad-Hoc Drag

Ad-hoc questions don't go away. They just pile up.

By enabling natural-language querying and exploration, teams reduce friction for stakeholders while dramatically cutting interrupt-driven work.

The result:

• Faster answers
• Fewer tickets
• Better stakeholder confidence

3. Automation as a First-Class Investment

One subtle but powerful change:

Dedicate ~15% of every sprint to automating repetitive manual work.

Not "when we have time."
Not "after the roadmap clears."
By design.

Over time, this compounds:

• Fewer manual fixes
• More stable pipelines
• Engineers working on higher-leverage problems

The 90-Day Framework I Use Consistently

Most data transformations don't fail because of technology. They fail because teams move too fast without alignment—or too slow without momentum.

The first 90 days matter more than any tool choice.

Days 1–30: Stabilize & Align

Focus: Stop the bleeding and align on outcomes.

• Meet stakeholders across Product, Engineering, and Business
• Define what "success" actually means (not just deliverables)
• Establish a shared north star: trusted, self-service data with clear ownership
• Baseline architecture, skills, risks, and data quality gaps
• Fix the most painful breakpoints to create breathing room

Outcome: Calm the system. Align expectations. Buy credibility.

Days 31–60: Build the Foundation

Focus: Create trust in the platform.

• Design a modern lakehouse that supports BI, analytics, and data science
• Model only critical domains—not everything
• Introduce a semantic layer so metrics are defined once and reused everywhere
• Embed data quality checks, lineage, and observability into pipelines
• Align Data Engineering and Data Science around shared outcomes

Outcome: Stakeholders stop questioning the numbers—and start using them.

Days 61–90: Deliver & Scale

Focus: Prove value and operationalize.

• Replace sprawling legacy reports with a smaller, outcome-aligned analytics suite
• Enable faster insight creation without breaking governance
• Launch executive scorecards focused on impact, not vanity metrics
• Establish operating rhythms: demos, retros, and data governance councils
• Shift culture from "reporting as an afterthought" to "data as a product"

Outcome: Momentum becomes self-reinforcing.

What Success Actually Looks Like

By the end of this phase, the win isn't just better dashboards.

It's this:

• New insights delivered in days instead of months
• Consistent metrics across teams
• Data teams aligned around shared outcomes
• Leadership confidence restored
• A platform that's ready for AI—not scrambling to catch up