Strategy must distinguish the preferred future state from the most valuable next move.
Executives often ask technology teams to identify the best long-term solution and then treat the answer as an investment sequence.
That is a category error.
A destination technology may depend on feedstock, infrastructure, regulation, skills or market conditions that do not yet exist at sufficient scale. Attempting to jump directly to the destination can destroy value. Refusing to move until every condition is ideal can be equally damaging.
The leadership problem is therefore not simply technology selection.
It is transition architecture.
The Strategic Context
Hu and Zhang's 2017 modelling of energy-conservation strategies for a Chinese steel producer provides a useful historical case.
The researchers compared three approaches: outsourcing coking, reducing the iron-to-steel ratio, and replacing the traditional blast-furnace/basic-oxygen-furnace route with direct reduction/electric arc furnace technology.
Within their model, DR/EAF was the strongest long-term option for energy and cost reduction. Yet the authors also identified a major constraint: the availability of scrap. Much steel embedded in buildings and infrastructure would not return to the scrap system for decades, making rapid substitution difficult under the conditions they analysed.
The paper therefore reached a strategically interesting position. The technically preferred future state was not necessarily the most feasible immediate move.
That is exactly the kind of distinction executives need to make in major industrial transitions.
What Leaders Commonly Misread
The first misread is technology ranking without dependency analysis.
A technology can appear superior on energy, operating cost or emissions and still be strategically premature if it requires unavailable inputs, grid capacity, transport infrastructure, specialist capability or regulatory approval.
The second is transition measures mistaken for end-state solutions.
The steel paper found coking outsourcing immediately feasible in its context and calculated attractive plant-level reductions. But it also acknowledged that outsourcing largely shifted energy consumption and cost to another industry rather than eliminating them from the total system.
A bridge can be useful without being the destination.
The third is assuming the transition will be linear. Real industrial systems contain asset lives, maintenance cycles, supplier dependencies, uncertain demand, learning curves and policy changes. The rational pathway may involve parallel technologies, temporary compromises and deliberate retention of optionality.
Reframing the Issue
Instead of asking, "Which technology wins?", leadership should ask four questions:
- What is the desired end-state capability?
- What conditions must exist for that state to perform as expected?
- Which of those conditions are under our control?
- What should we invest in now to improve current performance while increasing future option value?
This reframing converts a technology roadmap into a portfolio strategy.
It also prevents the organisation from confusing an engineering optimum with an enterprise optimum.
Destination Architecture Versus Transition Architecture
A destination architecture describes the future operating model the organisation believes will eventually be superior.
A transition architecture describes how to reach it without exceeding the organisation's capital, operational and capability constraints.
These are related but different.
For example, a manufacturer may conclude that electrification is the long-term direction. The transition architecture must still decide whether the next dollar should be spent on the electric asset itself, grid connection, energy storage, data capability, workforce training, supplier development or extending the life of an incumbent asset until enabling infrastructure improves.
The highest-value next investment may be an enabler rather than the headline technology.
Constraints Determine Sequence
The steel study places scrap availability at the centre of the DR/EAF transition. This is a classic systems constraint.
When a required input is scarce, investing in downstream conversion capacity can create under-utilised assets or expensive dependence on external supply.
The same logic applies elsewhere.
A hydrogen project without reliable low-emissions hydrogen is not a hydrogen transition; it is a dependency exposure.
A fleet electrification project without charging capacity is not a vehicle strategy; it is an infrastructure bottleneck.
An AI program without reliable enterprise data is not an intelligence program; it is an error-amplification risk.
A circular manufacturing model without reverse logistics is not circularity; it is an aspiration.
The constraint should therefore shape the sequence.
Outsourcing Can Improve the Boundary and Worsen the Decision
Hu and Zhang's coking example is particularly useful because the authors explicitly recognised the difference between company-level and system-level performance. Their model suggested that complete outsourcing could materially reduce the steel plant's reported energy use and cost, while total energy consumption and cost were largely shifted elsewhere.
This is why make-or-buy decisions should not be judged only by what disappears from the buyer's cost centre or emissions inventory.
Outsourcing may still be rational. Independent specialists may operate at greater scale, achieve better control or release capital from a non-core activity. But the decision should state which benefit is being pursued.
If the claimed benefit is system decarbonisation, displacement is not reduction.
Related article: When Efficiency Improves but Total Emissions Still Rise
Intermediate Measures Need Exit Logic
A compromise technology can create value if it has an explicit role in the transition.
The steel paper treated a lower iron-to-steel ratio as an intermediate approach: meaningful improvement without the full dependence of the end-state route.
This suggests a useful discipline for transitional investments.
Every bridge investment should state:
- what constraint it addresses now;
- how long it is expected to remain valuable;
- what future state it enables;
- what would cause it to be retired early;
- whether it creates lock-in that could delay the preferred future.
Without exit logic, temporary solutions become permanent through sunk cost.
The Capital Allocation Challenge
Industrial transition portfolios usually contain three kinds of investment.
1. Performance investments
These improve current operations: efficiency, process control, maintenance, heat recovery, yield, quality or energy management.
2. Enabling investments
These create conditions for future technology: grid connection, scrap systems, renewable supply, hydrogen infrastructure, data platforms, skills, supplier capability or standards.
3. Replacement investments
These install the future-state asset or process itself.
The temptation is to privilege replacement projects because they are visible and strategic. But the absence of enablers can make replacement premature.
Portfolio governance should therefore evaluate the dependency-adjusted value of each initiative, not just its stand-alone business case.
Decision Framework
A practical transition decision can be structured through seven tests.
1. End-state superiority
Why is the destination technology expected to be better? Consider total cost, emissions, throughput, quality, resilience and strategic fit.
2. Dependency readiness
List critical dependencies and score their availability, reliability and control.
3. Asset timing
Identify which incumbent assets are near natural replacement points and which still have significant economic life.
4. Reversibility
Assess whether the investment keeps future choices open or creates lock-in.
5. System boundary
Test whether benefits are eliminated, transferred or externalised.
6. Learning value
Determine whether an early pilot creates knowledge that materially improves later capital decisions.
7. Portfolio interaction
Assess competition for capital, shutdown windows, engineering resources and organisational absorption capacity.
The output should not be a single "go/no-go" answer. It should define a staged pathway.
From Strategy to Execution
Immediate action: identify the top three constraints to the preferred future state. Do not allow the technology roadmap to proceed without named owners for those dependencies.
Medium-term capability building: create a transition portfolio that separates performance, enabling and replacement investments. Link funding gates to evidence about dependency readiness rather than fixed dates alone.
Long-term strategic positioning: build option value. Secure supply relationships, develop technical skills, shape standards, pilot alternative technologies and avoid over-investing in assets whose economics depend on assumptions that may not survive the transition.
Where appropriate, establish trigger points such as:
- scrap availability reaching a defined range;
- renewable electricity contracts becoming reliable enough;
- infrastructure capacity becoming available;
- technology performance reaching a proven threshold;
- policy or market conditions changing project economics.
The roadmap then becomes conditional and adaptive rather than merely chronological.
Signals to Monitor
Monitor the variables that could change the transition sequence:
- availability and price of critical feedstocks;
- energy prices and grid intensity;
- supplier concentration;
- infrastructure lead times;
- technology maturity and failure rates;
- policy and carbon-cost changes;
- competing demand for the same resources;
- remaining life of incumbent assets;
- learning from pilots;
- evidence that an interim measure is becoming a source of lock-in.
The most dangerous signal is a roadmap that remains unchanged while its assumptions change.
Questions for the Leadership Team
- What is our preferred future operating state, and what evidence supports it?
- Which dependency most constrains our ability to move toward that state?
- Are we funding enabling capabilities at the same seriousness as headline technology?
- Which interim investments create option value, and which create lock-in?
- Are any apparent emissions or cost reductions simply transfers outside our boundary?
- Which current assets should be retired, extended or deliberately run down?
- What trigger would justify accelerating or slowing the transition?
Closing Perspective
Industrial transformation is rarely a choice between the old system and the perfect new one.
Leadership must manage the space between them.
The strongest strategy identifies the desired destination, exposes the constraints that make it difficult, and invests in a sequence that improves current performance while making the future state increasingly executable.
The best technology can be strategically wrong today and strategically necessary tomorrow.
The job of leadership is to know the difference.
Source basis: This article is an original ERANORTH synthesis principally informed by Rui Hu and Chao Zhang (2017), Discussion on energy conservation strategies for steel industry: Based on a Chinese firm, Journal of Cleaner Production, 166, 66–80.
About EraNorth Insights
EraNorth Insights publishes practical analysis on strategy, projects, operations, transformation and decision intelligence for professional and organisational use. About EraNorth.