Is developing a digital transformation strategy still worth it?

And yet, somewhere in the middle of that avalanche of capital, between seventy and eighty percent of digital transformation initiatives are failing to deliver what the original slide deck promised. Bain has put the number as high as eighty-eight percent.

That does not mean seventy, eighty, or eighty-eight percent of the money disappears. It means the initiatives themselves fail to meet their intended objectives, whether those objectives involve adoption, productivity, customer experience, operating cost, revenue, or some combination of all of them. The distinction matters, especially when the spending figures are large enough to invite lazy conclusions.

So when a CEO, a CFO, or a board member leans across the table and asks me whether developing a digital transformation strategy is still worth the trouble, my answer is rarely a clean yes or no. It is a question back at them: do you actually want a strategy, or do you just want a procurement plan dressed up as one?

Let me translate this for you. The strategy is not the problem. The hubris is. Anyone who has spent a decade in a corner office — or, like me, has watched the sausage get made from the next room over — recognizes the pattern instantly. Theorists in glossy suits sell the future. Middle managers procure the tech stack. The workforce gets the FOMO. And the spreadsheet eventually tells the truth: the money walked out the door while the value stayed in the pitch deck.

The $4 Trillion Paradox: Why Spending Isn't Delivering Results

Walk into any large enterprise right now and ask the chief technology officer what the organization is spending on. You will get a confident answer involving cloud migrations, AI pilots, data lakes, automation platforms, low-code tools, and probably a generative AI wrapper that has been rebranded three times since January.

The number they quote — if they are honest — will make your eyes water. Global digital transformation spending is projected to reach roughly $3.4 trillion by 2026 and around $4 trillion by 2027, accounting for more than two-thirds of information and communications technology costs globally.

That is not a small experiment. It is an enormous allocation of corporate and public-sector capital. But the size of the market does not tell us whether the spending is effective, and the failure rate for initiatives does not prove that most of the spending itself is wasted. It tells us something narrower and more useful: organizations are repeatedly approving projects whose intended outcomes are not being achieved or cannot be demonstrated convincingly.

That is a governance failure before it is a technology failure.

I watched this happen in 2008 with ERP rollouts. I watched it happen in 2014 with mobile-first initiatives. I am watching it happen now with AI. The pattern remains depressingly consistent: organizations spend lavishly on the technology, treat change management as an afterthought, and then act surprised when the ROI refuses to materialize.

Roughly seventy-three percent of organizations cannot definitively prove the ROI on their digital transformation investments. That does not mean every one of those investments has produced no value. It means the value is not being measured, attributed, or communicated with enough confidence to survive scrutiny. Trillions in projected spend, and a large majority of companies still cannot tell you with a straight face whether the investment moved the needle.

The strategy did not fail. The discipline did.

The friction is not necessarily in the technology. It is in the absence of a coherent framework that connects investment to outcome. Without a roadmap that ties technology decisions to business objectives, you are not transforming anything. You are placing bets. Some of those bets may pay off. Others may produce local improvements that never scale. Still others may solve a problem that stopped mattering halfway through implementation.

A serious digital transformation roadmap has to make those distinctions visible. It should identify the business constraint, the capability being built, the owner responsible for adoption, and the evidence that will justify the next tranche of funding. It should also make it possible to stop. That last part is where many strategies quietly become wish lists.

The project often survives because nobody wants to be the executive who cancels it. The business case gets revised, the deadline moves, and the definition of success becomes more forgiving. By the time the program is officially declared complete, the organization has learned to live with the gap between what was promised and what was delivered.

That is how spending continues while confidence disappears.

Beyond the Tech Stack: The Governance and Culture Multiplier

Here is the part the vendor decks never tell you. Companies that prioritize organizational culture and change management achieve up to 5.3 times higher digital transformation success rates than organizations fixated on the tech stack alone.

This is not a software problem. It is a human problem. It is a governance problem. It is a leadership problem involving people who do not want to do the unglamorous work of getting employees to use the new system, change the workflow, abandon the legacy process, and admit that the last fifteen years of transformation amounted to a series of half-implemented experiments.

I have sat in boardrooms where the CFO presents a beautiful slide on cloud migration cost savings and the CHRO has not been consulted. I have watched chief data officers get hired, handed a budget, and quietly marginalized because no one wanted to break the silos the new platform would expose. That is not strategy. That is theater with a capital request attached.

The cultural dimension is often described too softly, as if it were a matter of sending a few newsletters before go-live. It is more consequential than that. A new system changes who can make a decision, which team owns a process, what information is visible, and whose performance can be compared with whose. Resistance is not always irrational. Sometimes it is a warning that the transformation has redistributed risk without explaining the benefit.

That is why overcoming digital transformation resistance cannot be reduced to motivational messaging. People need to know what is changing, why it is changing, how their work will be evaluated, and what happens when the new process creates friction. Managers need authority to resolve exceptions. Executives need to reinforce the new operating model after the launch presentation is over.

Let me be clear: I am not anti-tech. I am anti-hubris. The technology is often fine. What is broken is the leadership theater around it.

A genuine digital strategy framework — the kind actually worth developing — begins with governance, not technology procurement. It defines who owns what, how decisions get made, what success looks like, and, most critically, what gets cut when the budget tightens. It also makes accountability cross-functional. A transformation owned exclusively by IT may deliver a system. It cannot, on its own, guarantee that finance, operations, sales, compliance, and frontline teams will change how they work.

The table above should look obvious. It is. And yet the industry keeps doing the wrong thing because the wrong thing is what gets funded. Technology purchases are tangible. Governance is not. A license can be approved, assigned, and depreciated. A change in decision rights requires executives to give up control, and that is a much harder line item to sell.

The strongest business digital transformation steps therefore start before the first platform is selected:

1. Define the business problem in operational and financial terms.

2. Assign ownership across the functions that must change, not only the team that will implement the tool.

3. Establish the baseline against which improvement will be measured.

4. Decide which legacy processes, applications, or reports will be retired.

5. Build adoption and accountability into the program rather than adding them after resistance appears.

6. Set a point at which the initiative will be expanded, redesigned, or stopped.

This is less exciting than a launch event. It is also where the value is either protected or lost.

The Data Quality Bottleneck in Modern Enterprise Architecture

If there is one friction point that quietly kills more transformation projects than almost any other, it is data quality. According to industry surveys, sixty-four percent of technology leaders cite data quality as their top operational challenge.

That is not a tooling problem. It is a discipline problem.

Let me describe what this looks like in practice. A company invests years and substantial capital standing up a data lake. The lake fills up — with garbage. Customer records are duplicated across multiple systems. Transactional data sits in incompatible schemas. Master data is managed by several teams using several different definitions of customer. By the time the analytics layer goes live, every report contradicts every other report, and the executive committee quietly loses faith in the numbers entirely.

The infrastructure may be modern. The data may still be unusable.

This is what happens when you treat digital transformation as an IT project rather than a business discipline. You can license the most expensive data platform on earth. If you have not done the foundational work of defining data ownership, establishing data contracts, documenting lineage, and enforcing quality standards at the source, you are building a mansion on quicksand.

Your data is your strategy. If it is a mess, your strategy is a mirage.

The problem becomes even more expensive when poor data quality is hidden behind polished dashboards. A dashboard can make inconsistent definitions look authoritative. An AI model can process bad inputs faster than a human team ever could. Automation can spread an error across an entire workflow before anyone notices that the original record was wrong.

That is why data quality has to be treated as part of enterprise architecture, not as a cleanup project assigned to analysts when the reports fail. The organization needs clear answers to basic questions:

• Which system is authoritative for each important data element?
• Who is accountable when the data is incomplete or contradictory?
• Which definitions must be shared across functions?
• Where should quality checks occur — at entry, during integration, or before use?
• What level of imperfection is acceptable for a particular decision?
• How will data issues be surfaced and corrected without creating a parallel bureaucracy?

Developing a digital transformation strategy without answering those questions is like hiring a Formula 1 driver before paving the road. The expensive part comes later, and it is the part nobody budgeted for.

Data quality also exposes a common weakness in digital transformation roadmaps: the tendency to describe capabilities without describing operating responsibilities. “Real-time analytics” sounds like a capability. It is not a complete plan. Someone has to define what real time means, which decisions depend on it, who acts on the signal, and what happens when the signal conflicts with an established process.

Until those decisions are made, the data platform is simply waiting for a business model.

Bridging the ROI Gap: Moving From AI Pilots to EBIT Impact

Now let us talk about the elephant currently standing in every boardroom: AI. Generative AI, to be specific.

Recent McKinsey research shows that nearly nine out of ten organizations now use AI in at least one business function. Yet only 5.5% report an EBIT impact greater than 5% attributable to AI.

That is a striking gap. But it needs to be read accurately. The 5.5% figure identifies the organizations reporting a substantial EBIT impact. It does not establish that every other organization is running a pilot, operating an AI center of excellence, or producing no income-statement impact whatsoever. Some may be capturing smaller gains. Some may be measuring poorly. Some may be building capabilities whose financial effect has not yet appeared. Others may indeed be stuck in experimentation.

The data does not tell us which category every organization belongs to. It does tell us that broad adoption is not the same thing as material financial impact.

The technology is real. The value capture is uneven. And the gap between the two has become one of the biggest sources of executive frustration in the market.

The issue is not that AI does not work. The issue is that many organizations have not built the operational scaffolding required to capture its value. They bought the tool without rebuilding the workflow. They deployed the model without redesigning decision rights. They trained the workforce without rethinking the processes that workforce actually performs.

In some cases, the AI deployment is not even aimed at a problem that matters financially. A team may automate a task that saves time but does not remove cost, increase throughput, improve conversion, reduce risk, or change the capacity of the business. The achievement is real. The business case is still weak.

A digital transformation strategy that bridges the ROI gap requires three things most companies skip:

1. A clear-eyed assessment of the process. Identify where AI can actually change cost, speed, quality, risk, revenue, or capacity rather than where it produces the most impressive demo.

2. End-to-end process redesign. Change the workflow around the technology, including the points where humans review, approve, correct, or take responsibility for the output.

3. A measurement framework tied to financial outcomes. Connect AI output to EBIT contribution or another decision-relevant business measure, not only to hours saved, prompts completed, or tasks automated.

There is a fourth requirement that tends to arrive after the excitement has faded: operating control. Someone must own model performance, data access, security, exceptions, and the consequences of an incorrect result. If nobody owns those questions, the organization will either overtrust the system or quietly stop using it.

The most useful AI roadmap is therefore not a list of use cases. It is a sequence of business decisions. Which process deserves intervention? What must change around it? How will the result be measured? What level of human oversight is appropriate? What evidence is required before the use case moves from experiment to standard operation?

Without those answers, AI becomes another line item in the opex budget. It may still be useful. It may even create local value. But nobody should confuse activity with transformation.

Reframing the Roadmap: Strategic Alignment Over Rapid Adoption

So, back to the original question. Is developing a digital transformation strategy still worth it in 2026?

My answer is yes — but only if you are willing to develop an actual strategy, not a procurement calendar with a glossy cover.

The roadmap is not obsolete. What is obsolete is the fantasy that you can skip the strategy and let technology vendors write it for you. Vendors are good at explaining what their products can do. They are not neutral arbiters of which problems your organization should solve, which capabilities it should build, or which projects it should abandon.

A useful digital transformation roadmap should be:

• Aligned to business outcomes, not vendor capabilities. If the roadmap reads like a product catalog, throw it out and start over. Begin with the constraint the business needs to remove.
• Honest about tradeoffs, including what you are going to stop doing. Every transformation budget should include a list of activities, applications, and reports being discontinued, not just things being added.
• Sequenced by leverage, not by hype. Tackle the highest-value, lowest-friction opportunities first when that sequence makes sense. AI is not always the answer. Sometimes the answer is consolidating the software tools you already pay for and few people use.
• Built around governance, not technology. Who decides? Who owns the result? Who measures it? Who resolves disputes between functions? Who has the authority to stop a project that no longer makes sense?
• Designed for learning. A roadmap should contain points where evidence can change the plan. A strategy that cannot be revised is not disciplined; it is merely stubborn.
• Connected to the operating model. If the future-state process is unclear, the technology investment is premature. Transformation happens when the organization works differently, not when a new platform appears in the architecture diagram.

The market will continue to spend at extraordinary levels. Trillions more will flow into cloud, AI, automation, and whatever the next buzzword ends up being. The available evidence does not justify claiming that most of this money will be wasted. It does justify a more uncomfortable conclusion: large spending does not protect an organization from weak prioritization, poor measurement, or a failure to change how work gets done.

Some initiatives will deliver. Some will produce partial value. Some will fail. The strategic question is whether the organization can tell the difference early enough to allocate capital intelligently.

The strategy is not the work. The strategy is what tells you which work is worth doing.

The companies that win will not necessarily be the ones that spend the most. They will be the ones that spend deliberately, with a clear link between capital, capability, adoption, and result. They will treat the roadmap as a financial instrument — something that allocates capital against expected returns, with discipline and accountability built into every line.

They will leverage the technology, certainly. But they will never confuse procurement for progress.

Here is the line I want you to walk away with: in digital transformation, the technology is often the easy part. The hard part is deciding what you are actually trying to accomplish — and having the courage to say no to everything that does not serve that goal. Anyone can buy a platform. Almost no one can build the discipline to use it well.

The $4 trillion question was never whether the money would get spent. It will. The question is whether your organization can connect its spending to outcomes, recognize when an initiative is not working, and redirect the investment before the slide deck becomes more valuable than the result.

Developing a digital transformation strategy is still worth it. But the strategy has to do more than describe the future. It has to decide what the organization will fund, what it will change, how it will measure progress, and what it will refuse to pursue.

The strategy was always the point.