IT Strategy Consulting: Could Your Team Start Monday From It?

IT strategy consulting helps leadership decide where technology money and people should go over the next few years, and in what order. A good IT strategy is one your team could start executing on Monday.

Last updated: October 22, 2026

There is a specific document that shows up in most enterprises about once every three years. Forty slides, a current-state assessment nobody disputes, a target-state architecture nobody argues with, and a three-horizon roadmap. It costs several hundred thousand dollars and it is technically correct. Twelve months later nothing in it has shipped.

The problem is not the analysis. It is that the document was never executable, and nobody applied the one test that would have caught it.

What is IT strategy consulting?

IT strategy consulting is outside help deciding where a company should spend its technology money and effort, in what order, and what it should stop doing to afford that.

Done properly it produces four things, and a strategy missing any of them is a point of view rather than a plan.

• A ranked set of decisions, not a list of initiatives.
• A named owner per decision, on the business side, not just in technology.
• A first shipped outcome per phase, with a date.
• An explicit stop list. What gets defunded to pay for this.

Notice what is not on that list: a target-state architecture diagram. Those are useful and they are not a strategy. A diagram describes a destination, and every organization we have worked in already knew roughly where it wanted to go. What it did not have was an agreed order and a name against each step.

The Monday test

Here is the whole filter, and you can apply it to a document you already own.

**Could a competent team start work on Monday using only this document, without asking the authors a question?**

If the answer is no, you bought analysis. Run through the specific failures it catches:

• No owner. A phase assigned to “IT and Operations” is assigned to nobody.
• No first outcome. “Modernize the data platform” is a direction. “One governed customer data product that the pricing team reads from, by end of Q2” is a phase.
• No stated dependency. If phase three assumes a capability phase one did not build, the plan is a wish sequenced as a schedule.
• No stop list. A strategy that only adds is a budget request wearing a strategy costume.
• No measure defined before the work. A baseline captured after the fact is not a baseline.

We apply this to our own documents, and it is uncomfortable in the right way. It is much easier to write a compelling three-year narrative than to name who is accountable for the first ninety days.

What an executable strategy contains

The caveat: the executable column is harder to sell and harder to sign off, because it forces people to say no to things in the room. A document that defers seven initiatives will get argued with. That argument is the value.

The one number that anchors the whole conversation

Before debating priorities, find out what share of your technology spend goes to keeping things running versus changing them.

The public sector version of this number is stark. In its July 17, 2025 report on modernizing critical decades-old legacy systems, the GAO noted that agencies “have typically reported spending about 80 percent on operations and maintenance of existing IT.” Large enterprises are not federal agencies, but the shape of the problem is the same: the run cost is the thing quietly setting your strategy, and it does it without a meeting.

That number reframes the exercise. A strategy that adds initiatives without touching the run side is competing for the twenty percent. A strategy that moves five points from run to change has funded itself. This is why legacy application modernization and enterprise modernization keep appearing in strategies that work, and why cloud migration scoped to a specific workload beats cloud migration as a program.

Where governance belongs, and where it does not

Governance has a bad name in technology organizations because it usually arrives as a gate rather than a decision structure. Worth noting that the standards bodies have moved on this point. When NIST released Cybersecurity Framework 2.0 on February 26, 2024, the significant structural change was adding a sixth function, Govern, alongside Identify, Protect, Detect, Respond, and Recover, covering how strategy, roles, policy, and oversight are established and monitored.

The useful read for an IT strategy is that governance is a function, not a checkpoint. In practice that means two things in the document: every decision names who makes it, and the strategy states what it will stop doing when the evidence says it is wrong. A plan with no mechanism for being disproven does not get changed, it gets abandoned.

What outside help is genuinely good for

An honest list, because most of the pitch in this category overstates it.

**Deciding faster.** The value is rarely a fact the organization did not have. It is a decision that internal politics could not reach.

**Naming the constraint out loud.** An outsider can say that the data is the problem, or that a vendor relationship is the problem, when an employee’s career makes that expensive to say.

**Building the first phase, not just describing it.** This is the part most of the category skips, and it is where a strategy either becomes real or becomes shelf-ware. We architect it, build it with your team, and hand over the playbook, which is a different engagement shape from writing a document and leaving.

Where outside help is weak: knowing your customers, your operational reality, and your political landscape. A consultant who tells you they have that on day one is guessing.

For the version of this scoped to a specific transformation program rather than the whole portfolio, see how we approach digital transformation consulting and how we structure a digital transformation roadmap.

How to tell the strategy is working

• Run versus change ratio. The number that funds everything else. Track it quarterly.
• Decisions made per quarter. Not initiatives launched. Decisions closed, including the ones to stop.
• Time from decision to first production change. The real measure of whether the strategy is connected to delivery.
• Number of things on the stop list that actually stopped. The most reliable predictor we know of whether a strategy is real.

Frequently asked questions

How long should an IT strategy engagement take?

Weeks, not quarters. If gathering the current state takes longer than shipping the first phase would, the engagement is the wrong shape. A focused team can evidence the constraints, rank the decisions, and name owners in four to six weeks, and anything longer is usually consensus-building being billed as analysis.

What is the difference between IT strategy and digital strategy?

In practice, the audience. IT strategy tends to cover the portfolio, the platform, and the run cost. Digital strategy tends to start from the customer or product and work back. The failure mode is identical in both, which is a document with no owners and no first shipped outcome, so apply the same test either way.

Should we hire a firm or build the capability internally?

Both, in that order. Use outside help to break a decision logjam and to build the first phase alongside your team, then keep the capability. If a firm’s engagement model depends on you not developing that capability, you will be renewing forever, which is the pattern most enterprises say they are trying to escape.

What if leadership will not agree on priorities?

Then the strategy is not the deliverable, the decision structure is. Name the two or three decisions that are genuinely contested, put the evidence for each on one page, and force a documented choice with an owner. A strategy built on unresolved disagreement will be relitigated at the first budget review.

Where to start

Take the last strategy document your organization paid for and run the Monday test on it. If a team could not start from it alone, you do not need a new strategy. You need the same content rewritten with owners, first outcomes, dependencies, and a stop list, which is a two-week job and a much cheaper one than the original.