Technology Is a Means, Not a Strategy

Technology creates strategic value only when leaders remain clear about the human, commercial and operational ends it is meant to serve.

A technology proposal can become difficult to challenge surprisingly early. Once it has a name, a sponsor, a budget estimate and a vendor presentation, the conversation often shifts from whether this is the right intervention to how quickly it can be implemented.

That is the point at which means begin to masquerade as ends.

A business does not need artificial intelligence because AI exists. It does not need automation because competitors are automating. It does not need a new enterprise platform because the existing system is old. These may all become legitimate investments, but only after leadership has defined the outcome, the constraint, the capability change and the consequences well enough to establish that the technology is actually the right means.

The strategic risk is not simply choosing bad technology. It is allowing technological momentum to determine organisational direction.

The Strategic Context

Stephan Lorenz's 2017 discussion of technological growth, drawing on pragmatist thinking, challenges the assumption that more technology is inherently equivalent to progress. His central concern is a dynamic in which technological means continue to multiply while the social and ecological ends they are supposed to serve receive less scrutiny.

The value of this argument for executives is broader than the environmental debate from which it emerges.

Modern organisations have strong structural incentives to add technology. New systems are visible. Projects have budgets. Platforms attract sponsors. Automation produces measurable activity. Digital capability can be shown on a roadmap. By contrast, deciding not to implement something, simplifying an operating model or improving managerial judgement can appear less tangible.

This creates an asymmetry in capital allocation. Technology has advocates, vendors, benchmarks and delivery plans. The alternative of not adding another layer often has none.

The result can be an organisation with more systems, more data, more automation and more technical capability, yet no corresponding improvement in customer value, decision quality, resilience or productivity.

What Leaders Commonly Misread

The first error is treating technological capability as strategic capability.

A business may possess advanced analytics without improving important decisions. It may automate a weak process and simply produce defects faster. It may deploy collaboration software while leaving accountabilities unresolved. It may introduce AI into knowledge work without deciding which judgements should remain human, which can be delegated, and how errors will be detected.

The second error is assuming that a technology is passive once acquired.

Technology changes the system around it. It modifies workflows, skills, incentives, data requirements, control points and sometimes the definition of good performance. A scheduling platform can change how managers allocate work. A predictive model can change which evidence receives attention. A self-service channel can shift effort from employees to customers. Automation can remove routine work while concentrating the remaining work into fewer, harder exceptions.

The means alters the end.

The third error is measuring adoption rather than value. User counts, licences activated, workflows automated and models deployed are implementation measures. They do not establish that the investment strengthened the enterprise.

Reframing the Issue

The strategic question is not, "What can this technology do?"

It is:

What must become materially better, and what combination of technology, process, capability and governance is most likely to make that happen?

That reframing moves the investment from a technology conversation to a system-design conversation.

It also changes the burden of proof. The sponsor is no longer required merely to demonstrate technical feasibility. The proposal must establish why the intervention is superior to realistic alternatives, what assumptions must hold, what new dependencies it creates and how leadership will know whether the intended outcome is actually emerging.

A useful discipline is to separate four things that frequently become blurred:

the end: the outcome that matters;
the means: the intervention selected to influence it;
the mechanism: why the intervention should produce the outcome;
the evidence: what would confirm or challenge that belief.

When any of these is unclear, confidence in the investment should fall.

Technology Creates Path Dependence

Some technology choices are more reversible than others.

A small experiment with a clearly bounded data set may be easy to stop. A core enterprise platform, proprietary architecture, deeply embedded automation layer or fleet-wide engineering standard can be much harder to reverse. The initial purchase decision may therefore commit the organisation to migration costs, specialist skills, vendor relationships, integration patterns and operating assumptions that persist for years.

This is why technology governance cannot be reduced to procurement approval.

The more difficult the decision is to reverse, the stronger the strategic evidence should be before commitment. Leaders should ask not only whether the technology works but what future options it closes.

A platform may create efficiency while reducing supplier flexibility. Automation may reduce direct labour while increasing dependence on scarce technical specialists. A cloud architecture may improve scalability while changing cyber, sovereignty or continuity risks. An AI system may increase throughput while making it harder to explain or audit important decisions.

There is no technology without a new risk architecture.

Capability Can Be Displaced, Not Just Added

Technology is often presented as a capability addition. Sometimes it is a capability substitution.

When people stop performing a task because a system performs it, the organisation may lose experiential knowledge that previously made exceptions visible. When decision support becomes highly automated, managers can become better at operating the interface while becoming less capable of independently testing the result.

This does not mean organisations should preserve inefficient manual work. It means leaders should deliberately decide which human capabilities remain strategically important.

For high-consequence decisions, the relevant question is not simply whether automation can reach an acceptable average performance. It is whether the organisation can recognise when the automated process is wrong, understand the failure and intervene before the consequence becomes material.

The retained capability may therefore be different from the old manual capability. The organisation may need fewer people performing the routine task, but stronger people able to challenge assumptions, interpret anomalies and govern the system.

The Portfolio Cost of Technological Enthusiasm

Technology decisions compete for more than money.

They consume scarce change capacity, executive attention, data-engineering effort, cyber review, subject-matter expertise, training time and operational tolerance for disruption.

A portfolio containing ten individually attractive technology projects can still be strategically poor if they depend on the same specialists, change the same processes or require the same business units to absorb multiple transitions simultaneously.

The relevant comparison is therefore not project versus no project. It is investment versus the best alternative use of organisational capacity.

This leads to a more demanding portfolio question:

If this initiative is funded, what will not be funded, delayed or adequately absorbed because of it?

If no one can answer, the opportunity cost is probably not being governed.

Decision Framework

ERANORTH recommends a six-test means-to-ends review before material technology commitments.

1. Define the outcome without naming the technology

Write the desired change in a way that would still make sense if the preferred technology disappeared tomorrow.

Weak: "Implement generative AI across customer service."

Stronger: "Reduce resolution time for repeatable enquiries while maintaining service quality, auditability and appropriate escalation for complex cases."

The second formulation leaves room for alternative means.

2. Identify the system constraint

Determine what currently prevents the outcome.

Is the constraint labour capacity, process variation, poor data, fragmented accountability, slow approvals, skill scarcity, physical throughput or something else?

Technology aimed at a non-constraint can produce local improvement without changing system performance.

3. Compare credible alternatives

At minimum, compare:

process redesign;
policy or governance change;
capability development;
technology augmentation;
automation;
outsourcing or partnership;
deliberate non-intervention.

The comparison should include value, time to value, reversibility, operational disruption, dependency risk and future option value.

4. Make the causal mechanism explicit

State why the proposed means should create the intended end.

For example:

"Automating data extraction will reduce manual preparation time, allowing analysts to spend more capacity on exception investigation."

This can then be tested. If analyst capacity is consumed elsewhere, the anticipated benefit may never materialise.

5. Define retained human judgement

Specify what the technology may decide, what it may recommend, what requires human approval and what independent checking remains necessary.

The higher the consequence of error, the more important this boundary becomes.

6. Establish stop and review conditions

Before implementation, define evidence that would cause the organisation to pause, redesign or stop.

Without pre-agreed challenge conditions, delivery momentum can become evidence in its own favour.

From Strategy to Execution

Immediate action should focus on reframing current technology proposals. Require each major initiative to state its intended outcome, constraint, mechanism, alternatives and reversibility before the next funding decision.

Medium-term capability building should strengthen technology portfolio governance. This means integrating architecture, cyber, finance, operations, customer impact and organisational change rather than treating each as a separate assurance stream. Benefits ownership should sit with the business leader responsible for the outcome, not only with the technology program.

Long-term strategic positioning requires building an organisation that can experiment without becoming captive to its experiments. Modular architectures, interoperable data, staged investment, strong internal technical judgement and disciplined exit criteria create option value. They allow the enterprise to learn while keeping future choices open.

Signals to Monitor

Leadership should watch for signs that means are displacing ends:

technology initiatives described primarily through features rather than outcomes;
adoption targets becoming the dominant success measure;
benefits with no operational owner;
increasing integration complexity without proportional value;
repeated requests for additional technology to solve problems created by earlier technology;
declining internal capability to challenge vendor or model outputs;
projects continuing because of sunk cost rather than current strategic merit;
a growing portfolio of "strategic" systems with no clear priority hierarchy.

The strongest warning sign is linguistic: when leaders can explain what is being implemented more clearly than why it matters.

Questions for the Leadership Team

What outcome are we pursuing that would remain valid if the proposed technology did not exist?
What is the actual system constraint, and what evidence shows this technology acts on it?
Which alternative interventions were seriously evaluated before the preferred solution was selected?
What new dependencies and irreversible commitments will this decision create?
Which human capabilities must remain strong after automation or augmentation?
What evidence would cause us to stop, redesign or reduce the investment?
What higher-value initiative might be delayed because this program consumes the same capital, people or executive attention?

Closing Perspective

Technology is neither the enemy nor the strategy.

The leadership responsibility is to keep means and ends connected as both evolve. A good technology decision strengthens the organisation's ability to produce valued outcomes, learn from consequences and retain future choice. A weak one accumulates capability without sufficient purpose and then relies on momentum to justify itself.

The difference is not technical sophistication. It is strategic discipline.

Source basis: This article is an original ERANORTH synthesis principally informed by Stephan Lorenz (2017), Ecological criticism of growth and the means and ends of technology: A pragmatist perspective on societal dynamics, Journal of Cleaner Production, 166, 98–106.

About EraNorth Insights
EraNorth Insights publishes practical analysis on strategy, projects, operations, transformation and decision intelligence for professional and organisational use. About EraNorth.