Move 02 · Understand the System

Fix the System,
Not the Symptom.

Digital experience problems are emergent. They arise from the interaction of identity, network, endpoint, application, and policy decisions made by teams that rarely sit in the same meeting. That's why ticket-by-ticket response never wins: every closed ticket treats a symptom while the system that produced it keeps producing more.

The Discipline

Five Ways of Thinking We Bring to Every Diagnostic

None of these are ours. They come from manufacturing, engineering, and decision science, and they've been proven for decades. What we bring is the practice of applying them to enterprise IT with real telemetry behind them.

Framework 01

First Principles Thinking

Strip away every inherited assumption and rebuild from what's actually true. Most IT operating models are accumulations: a service desk structure from the 2000s, an SLA regime from an outsourcing contract, a ticket taxonomy nobody remembers designing. First principles asks the only question that matters: what is IT actually for?

Our answer: keeping people productive. Not keeping servers up, not closing tickets fast. Those are means. When you rebuild the operating model from that principle, different priorities fall out naturally.

In practice: instead of asking "how do we reduce ticket volume," ask "how much productive time are employees losing, and where?" The first question optimizes the queue. The second one often reveals the queue is measuring the wrong thing entirely.
Framework 02

Systems Thinking

Experience is a property of the whole system, not of any component. A slow morning logon can involve the identity provider, VPN posture checks, group policy processing, disk encryption, and an aging device, each owned by a different team, each individually "green." No component owner can see the problem, because the problem only exists at the level of the interaction.

This is exactly why endpoint telemetry matters: the endpoint is the one place where all of those systems converge on a human being. Measure there, and the emergent problem becomes visible.

In practice: when five teams each report healthy dashboards and employees still complain, the diagnosis isn't that someone is lying. It's that nobody is measuring the seams. We instrument the seams.
Explore systems thinking applied to DEX in depth
Framework 03

Theory of Constraints

In any system, one constraint governs throughput. Improving anything other than the constraint is an illusion of progress. IT organizations routinely spread improvement effort across dozens of initiatives, most of which optimize non-constraints while the real bottleneck sits untouched.

Experience telemetry lets you find the constraint empirically instead of politically: the one issue, application, or policy that degrades the most productive time for the most people. Fix that, re-measure, find the next constraint. Repeat.

In practice: a prioritized issue backlog ranked by measured productivity impact almost never matches the backlog ranked by ticket volume or by which stakeholder is loudest. The gap between those two lists is where wasted improvement budget lives.
Framework 04

Inversion

Instead of asking "how do we design a great digital experience," invert the question: what reliably destroys one? Forced restarts during working hours. Logon delays. Applications that crash and lose work. Password resets that take a helpdesk call. Ticket ping-pong between resolver groups.

The destroyer list is finite, measurable, and rankable. Eliminating the top of it is usually worth more than any new capability you could add, and it's a far more tractable roadmap than chasing an abstract ideal.

In practice: we build the destroyer list from telemetry and sentiment, not from opinion. What employees name in surveys and what actually costs them time overlap less than most leaders expect. You need both sources.
Framework 05

Second-Order Thinking

Every IT policy is an experience decision, whether anyone treated it that way or not. Security hardening, software standardization, refresh deferrals, and support model consolidation all have first-order benefits that get measured and second-order experience costs that usually don't.

The point isn't that those policies are wrong. It's that the decision was made without seeing half the ledger. When the experience cost is measured, some policies survive scrutiny, some get tuned, and some turn out to cost far more in lost productivity than they save.

In practice: a posture check added to VPN logon is a security win and an experience cost. With telemetry, that cost becomes a number the CISO and CIO can weigh deliberately, instead of a slow drip of frustration nobody owns.
Every recurring ticket is the system telling you where it's broken. Closing the ticket answers the message without reading it.
One More Distinction

Not Every Problem Is the Same Kind of Problem

A subtle failure mode in IT organizations: applying the right method to the wrong class of problem. Borrowing from the Cynefin framework, two classes matter most in our work.

Complicated problems respond to analysis

A recurring application crash has a root cause. An expert with the right data can find it and fix it, permanently. These deserve root cause analysis, and telemetry makes that analysis fast.

Complex problems respond to probing

Why is Copilot adoption low in one business unit? Why do employees bypass the service desk? There is no single root cause to find. These require safe-to-fail experiments: measure, adjust, re-measure. Running root cause analysis on a complex problem produces a confident answer that's wrong.

The diagnostic tells you which one you're holding

Much of the value of an experienced practitioner is classification: knowing which problems will yield to a fix and which need an experiment. Getting this wrong wastes quarters.

Next · Move 03

A diagnosis without an operating rhythm decays into a slide deck.

Run It as an Operation
Get Started

What's the Constraint in Your Environment?

Most organizations can't answer that question with data. A diagnostic engagement changes that in weeks, not quarters.

Schedule a Conversation Back to Our Approach