A crack exists wherever the economics visible inside a decision or action differ from the economics ultimately borne by the enterprise.
Value leaks through six recurring mechanisms
- 01The enterprise can underwrite the wrong proposition.
- 02Individually defensible commitments can produce an inferior combined result.
- 03The investment can change while the approved economics remain attached to it.
- 04Installed capability can fail to become operating value.
- 05New evidence can change the economics without changing the commitment.
- 06The resulting estate can make the next modernization decision more expensive.
The residual state created by one generation of modernization changes the opportunity set, constraints, costs, and options facing the next.
Project performance is not enterprise return
Modernization begins as an economic promise.
The enterprise commits money, attention, expertise, operating capacity, organizational capacity, time, and future flexibility because it expects a better operating state later.
The return depends on the full commitment, not only the project that implements it.
The enterprise has to identify the right opportunity, compare materially different alternatives, allocate scarce resources among competing uses, account for interactions across the portfolio, carry the economics through implementation, convert installed capability into operating change, revise commitments when evidence changes, and leave a useful starting point for what follows.
A project can finish on time and still be a poor allocation of resources. A system can work correctly without producing the operating change assumed in its business case.
Positive project return does not establish that the enterprise chose the best investment. Reported return can also exclude costs pushed into later modernization.
Specialization creates handoffs
Complex enterprises divide difficult problems because specialization works.
Finance, operations, engineering, architecture, security, procurement, business units, and external specialists each see different parts of the economics.
Facts, assumptions, requirements, budgets, authority, technical decisions, operating constraints, and accountability then have to survive movement between them. Different participants can make sound decisions within their own mandates and still produce a poor enterprise result.
Sometimes the divergence begins with the underlying representation of the business.
Engagement
In one Talbot West engagement, a multinational telecommunications company had more than a dozen ERP environments inherited across acquired businesses. Core business concepts were represented differently across those environments.
Before forecasting and capital analysis could operate on a common enterprise view, those definitions had to be reconciled and encoded into a shared data architecture.
Each local environment could function adequately while the enterprise-level representation remained incomplete.
See the project: Enterprise data unification across acquired businesses.
Commitment makes these problems harder to correct. An assumption that is cheap to challenge when proposed becomes harder to revisit after it acquires a budget, architecture, provider, integrations, staffing, executive sponsorship, and dependent programs.
Time adds another problem. Reality continues changing while the enterprise executes the answer.
Executive teams, portfolio councils, enterprise architecture, PMOs, transformation offices, procurement functions, centers of excellence, and provider ecosystems close real gaps. Each still operates within a mandate.
The enterprise therefore has to preserve the economics across handoffs that cannot simply be removed.
Six ways modernization value leaks
Underwriting error
The enterprise commits against the wrong or incomplete economic proposition.
Before implementation begins, the enterprise has to establish the desired outcome and constraint, construct the opportunity set, compare materially different interventions, and decide which use of scarce resources deserves preference.
The opportunity can narrow too early.
- An ERP modernization directs attention toward systems.
- An AI initiative directs attention toward a technology class.
- A cloud migration establishes a destination.
Internal functions and external providers naturally see problems through the capabilities they possess.
A precise analysis of an incomplete opportunity set can still produce a poor investment. A weaker path can consume scarce capacity, delay learning, displace a stronger alternative, or close options that were never properly evaluated.
Engagement
In one Talbot West software engagement, a heavier AI architecture was tested against the behavior the product actually required.
A simpler architecture produced the required behavior with less computational burden, complexity, and control risk.
See the project: Architectural engineering for AI in legacy software.
Control
Keep the outcome, constraint, and proposed mechanism separate long enough to compare materially different alternatives and the resources each would displace.
Portfolio conflict
Individually defensible commitments can produce an inferior combined result.
Large enterprises make modernization decisions across business units, functions, technologies, providers, geographies, and budget cycles.
The enterprise inherits the combination.
- Procurement can lower acquisition cost while engineering absorbs higher integration cost.
- A business unit can avoid a migration while the enterprise inherits another bespoke environment.
- A security control can reduce one exposure while creating enough operating friction to produce workarounds.
- Several providers can make sound recommendations that depend on incompatible architectures or future operating states.
Programs also compete for architects, engineers, executive attention, operating downtime, data access, and organizational capacity for change.
One investment can enable another, make another unnecessary, consume a dependency, or foreclose a later path.
Portfolio decisions therefore have to account for shared constraints, dependencies, sequencing, integration burden, and the combined future created by separate commitments.
Project economics do not add cleanly into enterprise economics.
Translation drift
The approved economics remain while the investment changes.
A consequential commitment passes through many representations before it reaches operating reality.
- 01Strategic intent becomes requirements.
- 02Requirements become architecture.
- 03Architecture becomes program design.
- 04Program design becomes contracts, workstreams, backlogs, code, configuration, procedures, and operating behavior.
Each translation makes the decision usable by the next participant. Each can also remove context.
- A requirement can survive after its reason disappears.
- A procurement substitution can change an operating assumption.
- Scope can be reduced while most of the cost remains.
Ownership changes at the same time.
The executive who approved the economics does not write every requirement. Architects do not negotiate every commercial substitution. Engineering teams may receive instructions without the reasoning that made them economically sound.
The business case can remain unchanged while the thing consuming the capital becomes materially different.
Control
Material execution changes must remain traceable to the economic proposition they alter.
Conversion failure
Installed capability fails to become operating value.
Technology business cases usually assume changes beyond the technology.
The modeled return may depend on changed approvals, workflows, roles, staffing, controls, or retirement of legacy systems.
If the technical capability changes while the surrounding operating system remains largely intact, part of the modeled return remains stranded.
- Employees can use the new application while preserving the old spreadsheet.
- Automation can execute a task while the manual control remains.
- A new platform can operate while the legacy system remains licensed and supported.
Usage and economic conversion are different measures.
The investment case has to include the workflows, roles, decisions, incentives, controls, costs, and legacy mechanisms whose change is required for the expected return to exist.
Revision failure
New evidence changes the economics faster than the enterprise changes the commitment.
Execution produces information that did not exist when the decision was made.
- Integration proves harder than expected.
- Users behave differently.
- Actual operating costs become clearer.
- A competing mechanism improves.
The original decision can have been sound while the economics change.
The people who see the new evidence first often control implementation detail rather than the underlying capital allocation, architecture, provider, or premise.
Meanwhile, contracts, integrations, standards, staffing, executive sponsorship, dates, and dependent programs make revision progressively more expensive.
A program can remain green after the enterprise answer has turned red.
Control
The commitment needs explicit conditions for reconsideration, enough retained reversibility for reconsideration to matter, and a route by which material evidence can reach someone able to change the underlying decision.
Residual cost
The completed investment changes the economics of the next one.
Every modernization initiative leaves an estate: architecture, interfaces, data structures, contracts, provider dependencies, skills, workflows, governance, institutional knowledge, technical debt, workforce experience, and future options.
Some residual states compound capability.
Reusable data, cleaner interfaces, clarified operating logic, reduced uncertainty, and stronger institutional knowledge can make subsequent work cheaper or easier.
Others make the enterprise harder to change.
- Another platform creates another integration surface.
- Another semantic representation creates reconciliation work.
- Provider dependence raises exit cost.
- Specialized architecture requires scarce skills.
- Organizational fatigue reduces future change capacity.
Many of those costs arrive under another sponsor and another business case.
The residual estate then becomes part of the opportunity set and constraints presented to the next decision.
The original investment can retain its reported return while exporting part of its cost into the future.
Control
Evaluate the state the investment leaves under both success and failure, including future change cost, exit cost, reusable capability, retained knowledge, dependencies, organizational effects, and preserved options.
Each commitment changes the starting point for the next one
The six mechanisms are connected.
- 01Underwriting determines what receives resources.
- 02Portfolio interaction determines how the commitment affects other investments and shared constraints.
- 03Translation determines whether the approved economics survive implementation decisions.
- 04Conversion determines how much installed capability becomes operating value.
- 05Revision determines whether new information can still change the commitment.
- 06The residual state becomes part of the starting condition for the next decision.
Modernization therefore does not reset after each project. One generation of decisions changes the economics of the next.
How much of the promise survives?
Modernization yield describes how much of the original economic promise survives the full path.
- Some appears as realized financial benefit.
- Some survives as reusable operating, technical, informational, or organizational capability.
- Some appears in the options the enterprise preserves for what comes next.
No universal percentage combines those forms honestly. Different enterprises value capacity, knowledge, flexibility, and future options differently.
The financial arithmetic alone shows the leverage.
External evidence
One large transformation study illustrates the financial portion of the problem.
In McKinsey's 2021 global survey, respondents from companies reporting successful transformations estimated that they captured 67 percent of the maximum financial benefit available.
The same study attributed the lost value across the full transformation path:
If the available financial benefit is B, improving capture by p produces B × p of additional realized benefit.
- $3MThree additional percentage points against a $100 million benefit pool produces $3 million.
- $15MThe same three points against $500 million produces $15 million.
- $30MAgainst $1 billion, it produces $30 million.
That calculation excludes any effect retained capability or preserved options have on later decisions.
The investment case has to survive the handoffs
A capable finance organization, architecture function, PMO, operating organization, transformation office, executive team, and provider ecosystem can still lose the economic proposition as responsibility moves among them.
One consequential commitment has to remain legible enough to govern action from the original opportunity through the state the investment leaves behind.
The modernization continuity test
Take one consequential modernization commitment and trace it through six points.
- 01Underwriting
Can the enterprise reconstruct the outcome, constraint, materially different alternatives, expected economics, and assumptions that governed the original commitment?
- 02Portfolio
Are the commitment's interactions with other investments, shared constraints, dependencies, architecture, and scarce organizational capacity explicit?
- 03Translation
Can material implementation changes be traced to the assumptions and economics they alter?
- 04Conversion
Are the operating changes required to produce the expected return defined, owned, and measurable?
- 05Revision
Can material new evidence reach someone with authority to reopen the commitment while revision remains economically possible?
- 06Residual state
Before the investment is declared economically complete, has the enterprise accounted for what it changed about the costs, capabilities, dependencies, and options facing the next decision?
Excellent machinery at each individual point does not establish continuity between them.
The control objective
Complex enterprises will not eliminate modernization leakage.
- Specialization is necessary.
- Uncertainty is real.
- Conditions change.
- Execution produces information that did not exist when the original decision was made.
The objective is not perfect prediction or perfect coordination.
It is to preserve more of the original economics as a modernization commitment moves through the enterprise and changes the conditions inherited by what comes next.
At enterprise scale, even small improvements in modernization yield can be economically significant.
Bibliography
- Bucy, Michael, Bill Schaninger, Kate VanAkin, and Brooke Weddle. “Losing from Day One: Why Even Successful Transformations Fall Short.” McKinsey & Company, 2021.
- Brynjolfsson, Erik, and Lorin M. Hitt. “Beyond Computation: Information Technology, Organizational Transformation and Business Performance.” Journal of Economic Perspectives 14, no. 4, 2000.
- Flyvbjerg, Bent. “From Nobel Prize to Project Management: Getting Risks Right.” Project Management Journal 37, no. 3, 2006.
- Kruchten, Philippe, Robert L. Nord, and Ipek Ozkaya. “Technical Debt: From Metaphor to Theory and Practice.” IEEE Software 29, no. 6, 2012.
