Why Good Tech Decisions Still Fail
The decision was fine. That’s the part that stings. You did the work: compared the platforms, pushed back on the pricing, got the leadership team aligned, signed the contract. Eighteen months later, the system is half-adopted, two departments have quietly rebuilt their workflow in spreadsheets, and nobody can tell you whether the investment is earning anything back. Nothing went wrong on the day you decided. Everything went wrong in the year and a half that followed, which is the strongest argument for fractional IT leadership we know of.
What is fractional IT leadership?
Fractional IT leadership is an arrangement where a senior technology executive works with your organization part-time on an ongoing basis, providing CIO or CTO-level strategy, vendor governance, and board-level reporting without the cost of a full-time hire. Unlike a consultant brought in for a defined project, a fractional leader stays past the go-live date. That distinction matters more than the cost savings, because the period after implementation is where most technology value is either protected or quietly lost.
Most technology investments fail after the decision, not during it
The evidence points away from the decision itself. Boston Consulting Group’s 2020 study, “Flipping the Odds of Digital Transformation Success,” drawn from 825 senior executives and 70 transformation programs, found that only 30 per cent met or exceeded their target value and produced lasting change. Another 44 per cent created some value but fell short of their targets, with limited long-term change to show for it. Those organizations did not fail to choose. They chose, and then the choice stopped mattering.
Deloitte’s 2025 Chief Transformation Officer Study arrives at the same place from a different angle. Its strongest predictors of a successful program were sustained C-suite engagement and a dedicated leader accountable for the outcome, not the quality of the original business case.
Read those two findings together and an uncomfortable conclusion falls out. The effort you put into making the right call has a smaller effect on the result than the effort you put in afterward. Most mid-market organizations invest the opposite way.
Every technology decision moves through four phases
We work through a framework called Dream, Discover, Decide, Deliver. The reason it’s four phases rather than one is that organizations rarely fail evenly across all of them.
Dream is where ambition gets grounded in a business objective. What are we actually trying to change, and does the leadership team agree on what that means?
Discover is the honest assessment: the systems, contracts, vendors, and workflows that hold the environment together, or fail to. This is the stage where you find out what you’re really paying for.
Decide is the structured choice. Weighted scorecards, a managed RFP, facilitated demos, and a recommendation someone can defend to a board.
Deliver is everything after. Governance, vendor accountability, adoption, measurement, and the willingness to revisit a decision when the business moves.
Three of those phases have a natural end. Deliver doesn’t.
Deliver is the phase nobody staffs, because it never ends
Here’s why the Deliver phase gets skipped, and it isn’t negligence.
Dream, Discover, and Decide all have deadlines. They produce something you can hold. Someone can put them on a project plan, attach a budget, and close them out. Deliver has no completion date and no obvious owner. It shows up as an ongoing operating cost rather than a project line, which makes it the easiest thing in the world to defer.
So it gets absorbed instead. The IT manager picks up vendor management on top of an operational job that was already full. The CFO inherits contract renewals. The CEO fields escalations. Nobody owns the question of whether the investment is still the right one, because that question doesn’t belong to anybody in particular.
The Project Management Institute’s Pulse of the Profession research on benefits realization found that 83 per cent of organizations lack maturity in tracking whether projects deliver the benefits they promised. That figure dates to 2016, and nothing since suggests mid-market organizations have closed the gap. If anything, the number of vendor relationships a 200-person company manages has grown considerably.
Your vendor cannot govern your vendor relationship
This is the part that tends to land awkwardly in a room.
When the Deliver phase has no internal owner, it gets handed to the vendor by default. Your managed services provider runs the quarterly business review. Your software partner reports on adoption. Your implementation firm tells you how the implementation went.
None of those people are lying to you. They’re being asked to grade their own work, and they have a commercial interest in the answer. A QBR run by the vendor will surface the problems that vendor can profitably solve. It won’t surface the possibility that you’ve outgrown the platform, that two of your systems overlap heavily, or that the renewal sitting on your desk should be a re-tender.
Governance only works when the person doing it has nothing to sell you. That’s the reason our practice takes no referral fees and resells nothing.
Where fractional IT leadership actually earns its money
In practice, a fractional leader working the Deliver phase does work that looks unglamorous and compounds quietly:
- Running vendor reviews from your side of the table, contract open, asking whether the SLAs were met and whether the relationship still fits where the business is going.
- Watching adoption rather than deployment. A system that’s installed and a system that’s used are different outcomes, and only one of them shows up in the return.
- Keeping a live technology roadmap tied to the next 12 to 24 months of business objectives, so the next decision starts from evidence rather than a vendor’s renewal calendar.
- Translating technology performance into language a board can act on: what we spend, what we get, where the risk sits.
- Mentoring the internal IT team so capability stays in the building rather than walking out at the end of an engagement.
- Saying out loud when a decision the organization already made is no longer the right one. That’s the hardest thing to get from anyone with a stake in the outcome.
None of that requires 40 hours a week. All of it requires seniority and independence.
You cannot deliver what you never discovered
There’s a sequencing trap worth naming, because it’s the most expensive version of this mistake.
Organizations that feel the Deliver problem often try to jump straight to ongoing leadership. They bring in a fractional CIO and ask them to fix the governance. The fractional leader then spends the first two months doing discovery anyway, because you can’t govern an environment nobody has mapped, and there’s no current inventory of contracts, systems, spend, or stakeholder pain to work from.
That’s a reasonable use of the time. It’s also an expensive way to buy an assessment.
The cleaner path is to run a Discovery Sprint first. Two to four weeks, structured, with stakeholder interviews, a systems and vendor audit, and a written findings report with prioritized recommendations. You come out of it knowing what you have, what it costs, and what genuinely needs a decision. Then you can make an informed call about whether ongoing fractional leadership is warranted, and if it is, the person you bring in starts on day one with a map instead of a blank page.
Roughly 70 to 80 per cent of the Discovery Sprints we run lead to a follow-on engagement. The more useful reading of that number is the inverse: a meaningful share of organizations find out they don’t need what they came in asking for. That’s a good outcome too.
Signs your organization has skipped the Deliver phase
You don’t need a diagnostic for this. Any one of these is enough:
- You signed a significant technology contract in the last two years and nobody has formally reviewed whether it delivered what it promised.
- Your quarterly business reviews are prepared and presented by the vendor being reviewed.
- You can’t produce a current list of your technology contracts, their renewal dates, and their annual cost in under an hour.
- A system you paid to implement now has a spreadsheet workaround built around it.
- Your technology roadmap was last updated by someone who no longer works there.
- Renewals get approved because cancelling feels riskier than continuing, not because anyone assessed the value.
In every one of those cases, the work of sustaining the decision was never assigned to a person.
Where to start
If that list described your organization, the useful next step isn’t to go shopping for a fractional CIO. It’s to find out what you actually have, so that whatever comes next, whether ongoing fractional leadership, a structured vendor selection, or cleaning up a handful of contracts and moving on, is a decision made on evidence.
Book a discovery call, and we’ll help you work out which one you need
FAQ
How is fractional IT leadership different from hiring a consultant?
A consultant is engaged for a defined project with an end date and a deliverable. Fractional IT leadership is ongoing and embedded. The fractional leader attends your leadership meetings, accumulates context about your business over months and years, and stays accountable through implementation, governance, and the next decision cycle. A consultant helps you make one good decision. A fractional leader helps that decision keep paying off.
We already have an MSP. Isn't governance their job?
Your MSP governs its own delivery, which is a different thing from governing your technology portfolio. They have a commercial interest in the scope of their own contract, so they aren't the right party to assess whether that contract should shrink, expand, or go to market. Most mid-market organizations need both: an MSP running operations, and an independent voice holding the relationship to account.
How do I know if we skipped the Deliver phase?
Ask who is accountable for whether last year's technology investments actually worked. If the answer is a name, you're probably fine. If the answer is "a few of us keep an eye on it," or the room goes quiet, the phase was skipped.
Can we start with an assessment instead of committing to ongoing leadership?
Yes, and usually you should. A Discovery Sprint runs two to four weeks and ends with a written findings report, a technology roadmap or vendor scorecard, and a working session with your leadership team. It gives you the information to decide whether ongoing fractional leadership is worth the spend, rather than deciding blind.
How long should a fractional engagement run?
Long enough to survive one full cycle: a planning period, a decision, an implementation, and a couple of quarters of governance after go-live. Anything shorter tends to end right before the part that matters. Most engagements start with an initial three-month commitment and continue month to month from there.
When is the right time to engage Deliver Digital?
Ideally before selection begins. But we also help mid-project—when leaders realize what they bought isn’t what they needed. Either way, our goal is clarity, not complexity.
