The IKEA Effect of Digital Transformation
The core lesson: people value what they help build — but only when it actually works
The IKEA effect captures a simple but powerful truth: we tend to value things more when we’ve had a hand in making them. In the original research by Norton, Mochon, and Ariely, people valued self-assembled products more highly than identical pre-built ones.
But there was an important catch: the effect weakened—or even disappeared—when the work was unsuccessful, incomplete, or undone. Put simply: effort creates ownership only when it leads to something usable. (SSRN)
That’s a great lens for digital business transformation.
Transformation can’t just be “installed” like software. It has to be built by the business—with finance, operations, sales, supply chain, risk, technology, data, compliance, and frontline teams all contributing.
But, just like flat-pack furniture, if the screws are missing, the instructions contradict each other, and the parts come from three different boxes, excitement quickly turns into frustration.
That’s what happens when organisations rush into large-scale AI, analytics, automation, or platform modernisation while the underlying data foundation is still fragmented.
The business version of the IKEA effect
In a healthy transformation, business teams don’t feel like change is being done to them. They feel like they’re helping build something that actually solves their problems.
That matters because many transformation failures aren’t caused by weak technology. They’re caused by weak ownership, unclear process design, poor data quality, and fragmented execution.
Recent AI research keeps pointing in this direction. McKinsey’s 2025 global AI survey found that AI use is widespread, but many organisations still struggle to move from pilots to enterprise-level impact. High performers are more likely to redesign workflows, invest in data and technology infrastructure, and embed AI into business processes. (McKinsey & Company)
MIT Sloan makes a similar case: AI impact comes from redesigning work, workforce, and workplace together—not bolting AI onto old ways of working. (MIT Sloan)
So the IKEA-effect lesson is this:
Transformation succeeds when people help build the new system of work — but it only creates value when the “assembly” is complete, coherent, and usable.
Why the data foundation is the “floor” under the furniture
Many businesses want to start with the exciting part: AI copilots, predictive models, automated workflows, customer intelligence, self-service dashboards, digital twins, or agentic AI.
But if the organisation still has data silos, duplicate customer records, conflicting product definitions, manual reconciliations, broken handoffs, and unclear ownership, then deploying AI at scale is like building a showroom on an uneven floor. It might look impressive on launch day, but the cracks appear quickly.
A weak data foundation leads to predictable problems:
Sales says “customer” means the buying account.
Finance says “customer” means the bill-to entity.
Operations says “customer” means the ship-to location.
Marketing says “customer” means the individual contact.
Risk says “customer” means the legal entity.
Then an AI model gets asked: “Give us a single view of the customer.”
It can’t. All it can do is amplify the confusion.
That’s why McKinsey’s 2026 guidance on agentic AI stresses continuous data quality management, lineage, governance, access controls, and curated internal data as strategic differentiators before scaling AI agents. (McKinsey & Company)
The dangerous shortcut: confusing visible progress with real progress
The danger in transformation is that visible technology can create the illusion of progress.
A new dashboard looks like progress. A chatbot looks like progress. A cloud migration looks like progress. A new AI pilot looks like progress.
But if the same duplicated, poorly governed, inconsistent data is still flowing underneath, the organisation hasn’t transformed. It has just given old problems a modern interface.
BCG made a similar point in 2026: only a small share of companies are generating sustained P&L impact from AI, while many see negligible material benefit. Often the issue isn’t the technology—it’s weak transformation discipline and fragmented efforts. (BCG Global)
That’s the failed-IKEA scenario: people worked hard, but the cabinet still wobbles. The emotional result isn’t ownership—it’s cynicism.
Teams start saying:
“We tried transformation before.”
“The dashboard numbers are wrong.”
“The AI doesn’t understand our business.”
“IT built something we can’t use.”
“The process still breaks—just faster.”
Once that happens, the next transformation has to fight not just technical debt, but emotional debt.
The better lesson: let the business co-build the data foundation
The IKEA effect isn’t an argument for making everyone suffer through complexity. It’s an argument for meaningful participation.
In transformation terms, that means business domains help define, validate, and own the data that powers their work. Not in abstract governance meetings, but through practical questions like the following:
What is the authoritative source for customer, product, supplier, employee, asset, contract, and transaction data?
Which duplicates are acceptable, and which create business risk?
Which processes create bad data at the source?
Which handoffs trigger rekeying, reconciliation, or spreadsheet workarounds?
What do "done", "active", "profitable", "delivered", "resolved", or “compliant” mean across domains?
Which data needs to be real-time, and which just needs to be correct by the close of business?
Who owns the quality of each critical data element?
This is where cross-domain initiatives start to work in practice. They stop being technology projects and become shared truth-building exercises.
For example, an order-to-cash transformation isn’t just about automating invoices. It requires sales, finance, operations, logistics, customer service, and data teams to align on the facts that move across the value chain: customer identity, contract terms, pricing, credit status, order status, delivery confirmation, invoice accuracy, payment allocation, and dispute reason codes.
When those facts align, you can measure success:
fewer duplicate customer records
fewer invoice disputes
reduced manual reconciliations
faster cycle time from order to cash
improved forecast accuracy
cleaner handoffs between teams
higher trust in dashboards and AI outputs
reduced operational leakage
clearer accountability for data quality
That isn’t “data for data’s sake.” It’s business performance.
The vivid analogy: AI is the power tool, data is the material
AI is a power tool—but a power tool doesn’t fix warped wood, missing screws, or the wrong blueprint.
If the data is duplicated, AI will duplicate the confusion.
If the process is broken, AI will accelerate the breakdown.
If ownership is unclear, AI will expose the ambiguity.
If definitions conflict, AI will produce confident but contested answers.
If lineage is missing, no one will trust the result when it matters.
The organisation doesn’t need to delay innovation indefinitely. But it does need to avoid building the future on wet cement.
The leadership message
The IKEA effect reminds leaders that ownership is built through participation, not communication alone. People support what they help shape.
But the deeper lesson is sharper: participation has to lead to completion, usefulness, and visible value.
So the right sequence isn't
Buy technology → deploy solution → ask the business to adopt it.
It’s:
Clarify business outcomes → fix the data foundation → redesign cross-domain processes → co-build with users → scale technology where the facts are trusted.
That’s especially important before large-scale AI deployment. AI doesn’t remove the need for data discipline—it raises the cost of not having it.
The takeaway
The IKEA effect is a powerful reference for transformation because it captures both the emotional and operational truth of change:
People value what they help build, but only if what they build stands up.
A digital transformation “stands up” when the organisation has a reliable data foundation: common definitions, trusted sources, governed access, clear ownership, clean handoffs, and measurable business outcomes.
Without that foundation, AI becomes a shiny layer over organisational confusion. With it, AI becomes a force multiplier for better decisions, faster processes, and cross-domain business value.
The goal isn’t to build more technology. The goal is to build a business that can trust its facts.
