Despite heavy investment in data platforms, dashboards, and advanced analytics teams, many organizations still struggle to turn data into decisions. According to Dr. Wendy Lynch, CEO of Analytic Translator, the problem is not technical. It is conversational.
In a recent interview, she argued that the gap between business leaders and data teams is less about skill and more about how requests are made, interpreted, and refined, or more often, not refined at all.
The hidden breakdown: speed and unawareness
Dr. Lynch describes the most common failure point in analytics work as a shared blind spot between business and analytics teams.
“The most common breakdown is a combination of speed and unawareness, on both sides. Business leaders often don’t have the vocabulary to fully articulate what they need. They know something is wrong, or they sense an opportunity, but what comes out is a proxy for the real request — not the thing itself. On the analyst side, most are never taught to explore the request behind the request. They receive something that sounds specific enough, and they execute on it. Nobody is being careless. Nobody knows a better conversation was possible.”
The outcome is predictable. Requests become short, vague instructions paired with deadlines, and analysts spend weeks producing outputs that miss the actual need.
Lynch says the persistence of this pattern is not accidental.
“Only 5% of analytic projects deliver exactly what the business needs, first try, no rework.”
She stresses this is not a talent problem. It is a communication failure that happens before analysis even begins.
Senior leaders recognize the cost as well.
“In my interviews of CTOs and CAOs, they admitted that their teams are 50-80% less effective (and more stressed and demoralized) than they could be if communication between business teams and analytics teams were better.”
The missing skill: better discovery conversations
For Lynch, the solution starts with how work is framed at the very beginning.
She argues that organizations consistently undervalue discovery conversations and treat them as optional.
“Start with the discovery process — and treat it as the real work, not an extra step. What needs to happen before any analysis begins is a real conversation: invite context, reflect what you heard, surface the actual goal behind the stated request. This takes five minutes.”
She gives a simple example. A team once requested a full redesign of a customer dashboard with a three-week timeline. After a short conversation, the real need became clear.
“We learned the business needed more urgency for sales managers. More red on the dashboard. That change took five minutes. The discovery process didn’t cost time. It saved three weeks of it.”
The issue, she says, is that neither side is trained to have these conversations. Analysts are not taught to question intent, and business leaders are not taught how to fully express it.
“That skill — knowing how to have that conversation — can be taught. And that’s the practical step most organizations skip entirely: train your people in analytic translation. Not just analysts. Business leaders too. Both sides need a shared language for how to make a request, how to explore it, and how to confirm that what was heard is what was meant.”
Building structure into collaboration
Training alone is not enough. Lynch says organizations also need structural habits that reinforce better communication.
“The second practice is cultural: build translation checkpoints throughout every project, not just at delivery. Scheduled check-ins at design, mid-project, and before results land. Analysts learn to explore the request before touching the data. Business leaders learn to expect questions and stop treating them as incompetence or delay.”
This changes the role of questioning from friction to alignment.
Over time, she says, teams stop assuming misunderstanding is a problem and start recognizing it as a normal part of doing the work properly.
“When organizations invest in teaching these skills together — as a team capability, not an individual fix — the dynamic shifts. People stop assuming the other side is difficult and start recognizing they’ve just never been given the tools to bridge the gap.”
Her conclusion is blunt.
“None of this requires a new hire. It requires a decision that structured conversation is not a soft extra. It’s what determines whether any of the technical work ever gets used.”
What “good” alignment looks like
If organizations improve communication, Lynch says they need better ways to measure success.
She warns against relying on dashboards, usage metrics, or delivery speed alone.
“Measure how often both teams understand the purpose of the request (to inform a decision or action), and whether the information was used for that purpose. Don’t rely only on transactional measures like dashboard adoption and delivery timelines (lagging indicators). Leading indicators are simpler and harder to fake: does the business team use the results to make a different decision? Not ‘did they attend the presentation.’ Did anything change?”
She describes baseline reality as messy. Most projects require rework because initial requests are unclear or incomplete.
Improvement shows up gradually. Fewer iterations. Clearer requests. Stronger ownership of decisions.
“If requests are arriving clearer, rework is going down, and business leaders can articulate what they decided and why data was part of it — you’re closing the gap.”
But the strongest signal is cultural, not technical.
“What ‘good’ looks like in practice is almost mundane when you see it. Requests arrive with context because there’s enough trust that people say what they need. Analysts ask questions before starting and nobody treats that as delay. Presentations lead with what matters to the audience, not what was most interesting to the team. And when results are inconvenient — when the data doesn’t show what leadership hoped — someone says so clearly, without drama and without apology.”
That final point, she says, is the real indicator of maturity.
“That last one is the real tell. An organization where inconvenient findings can be stated plainly and still lead to action has done this work.”
The takeaway
Across Lynch’s perspective, the message is consistent. Most analytics failures are not failures of data or talent. They are failures of translation.
Until organizations treat conversation as a core skill, not a soft one, even the most advanced analytics teams will continue producing work that is technically correct but practically misaligned.
