Almost every organisation with a decarbonisation obligation has three things already: a target, at least one energy audit, and a list of technologies someone has recommended. Very few have a sequenced portfolio of projects that a capital committee would approve.
The distance between those two states is where programmes stall, and it is not primarily a technical distance. It is the work of turning an identified measure into a defined project: establishing a load basis from data rather than nameplate ratings, comparing options honestly, producing an estimate someone will stand behind, understanding what the change does to safety and compliance, and deciding in advance how the result will be proven.
The central argument
Decarbonisation programmes do not usually fail because the wrong technology was chosen. They fail because nobody did the engineering and commercial work required to make a decision defensible.
This matters because the two failure modes call for entirely different responses. If the problem were technology selection, the answer would be better technology assessment. If the problem is execution capability, more technology assessment makes it worse — it produces more options, none of them developed enough to fund.
What the gap looks like in practice
- An audit from three years ago identifying measures nobody costed, sequenced or assigned.
- A target published without a portfolio of projects capable of reaching it.
- Business cases returned by the capital committee for reasons the engineering team reads as bureaucratic and the finance team reads as basic.
- Enabling infrastructure — electrical capacity, distribution, metering — never funded, because it has poor standalone returns and nothing forces the dependency into view.
- Savings claimed at approval and never checked, so the next appraisal inherits assumptions nobody has tested.
Why it persists
The capability required sits between disciplines. It needs enough engineering to know whether a measure is safe and constructible on this site, and enough commercial discipline to know what a capital committee needs in order to say yes. Organisations tend to have one or the other.
Energy audit firms are efficient at identifying opportunity and usually stop before feasibility. Large consultancies can do the development work but are expensive for early-stage shaping. Technology vendors will develop a project enthusiastically, within the constraint that it uses their technology. Sustainability advisers operate at board level and may lack utility and process engineering depth.
None of these is a criticism. They are different businesses solving different problems. But it explains why the gap survives: there is no natural owner of the work, so it does not get done, and its absence is attributed to a shortage of good projects rather than a shortage of project development.
Practical implications
- Treat project development as a distinct capability to be resourced deliberately, not as something that happens between an audit and a contractor.
- Judge an audit by whether you could fund something from it, not by how many measures it lists.
- Fund enabling projects on a different argument from measures with direct returns. Putting them through the same hurdle rate guarantees they fail it and the programme stalls behind them.
- Decide how you will verify before you commit, because after commissioning the metering that would have made verification possible is no longer installable at reasonable cost.
Where this argument goes next
If the constraint is execution capability, the useful question becomes what that capability actually consists of. The chapters that follow take it apart: how a baseline earns trust, how options comparison stays honest, how maturity should govern the weight a figure carries, and how governance keeps a portfolio moving when capital is contested.
The next published summary — Evidence and maturity — deals with the one that causes most damage in practice: figures presented without the basis needed to judge them.