A carbon footprint without an action plan is a diagnosis without a treatment

The report stays on the table, closed, after everyone has left the room. It reopens eighteen months later, when somebody is asked to update the figures, and the number has gone up.

The company grew, which is the whole explanation. Nobody had been assigned to make the number do anything else. And the report itself was usually good: scopes correctly separated, methods documented, uncertainty acknowledged, the works. This is not a failure of sincerity, it is a failure of structure, and structure can be fixed.

Why good companies stop at the report

The measurement project ends in the very meeting where the reduction work should begin, and the room does not notice, because the meeting feels like a completion. Something was commissioned, delivered and presented. The consultant or the internal lead gets thanked. There is, at that moment, no next agenda item, and the diagnosis quietly becomes the deliverable.

Three forces hold companies there. The measurement had a project structure, a budget, a deadline and a person; the reduction has none of those yet, so it competes as a vague intention against everything in the company that does have a structure. The levers belong to people who were not in the room, because the biggest lines are purchasing decisions, fleet decisions and product decisions, owned by managers who saw the footprint as a presentation rather than as their objectives changing. And the next external prompt, the customer questionnaire, the rating, the BEGES cycle, asks whether you measured far more insistently than it asks whether you reduced.

That last point deserves to be said more bluntly than it usually is. The reporting ecosystem, taken alone, rewards measurement. You can pass every compliance gate of the next two years with a footprint that only grows, which is precisely why reduction needs the structure that measurement got for free.

Five lines, five names, five dates

The transition plans that work are embarrassingly short. Not a sustainability strategy document; a table.

Take the five largest emission sources from the footprint, which typically cover most of the total. For each one: the reduction lever chosen, the person who owns it, the date the first action is done, and the budget or arbitration it needs, signed. Add a sixth row for the measurement itself, whose cadence and next scope improvement also get an owner and a date, because the data quality ratchet belongs in the plan too.

Two properties separate this table from the strategy documents that fail. The owners are the operational managers of the lines, the purchasing director for the purchasing line, the fleet manager for the fleet, not the CSR manager, whose job is to run the table rather than to own every row. A footprint's levers live where its emissions live, in purchasing categories, on factory floors, in travel policies, and assigning them all to the person who measured is how plans die. And every row survives contact with a budget cycle: an action that needs money names the amount, an action that needs no money names the arbitration it does need, because "costless" actions that die in committee always turn out to have needed one.

Running the ninety-day review so it survives

Then hold the second meeting, ninety days later, same room, one question per row: what moved. That meeting, more than any document, separates the companies that reduce from the companies that remeasure. It also produces exactly what rating frameworks mean by actions and results, which a filed PDF does not.

It has its own failure mode, though: it degrades into a status ritual where every row is "in progress" and nobody is lying, exactly, but nothing is moving either. Three rules keep it honest.

Only three answers exist per row. Done, with the evidence named. Moved, with a new date and the reason the old one broke. Or blocked, with the arbitration it needs and who can make it. "In progress" is not an answer, it is the absence of one, and a chair who accepts it once will hear it five times per meeting thereafter. The discipline sounds harsh and is in practice a kindness, because it converts vague guilt into a specific request, which is the only form of trouble a management committee can actually resolve.

A stalled row gets one of three treatments, chosen out loud. Split it, because a row that has not moved in two cycles is usually two decisions wearing one name. Re-own it, because sometimes the honest news is that the named person cannot carry it beside their day job. Or close it deliberately, minuted as a decision not to act, with the emissions it leaves on the table stated. Closing a row consciously is respectable; letting it rot is what teaches everyone the table is decorative.

Keep the minutes as compliance evidence. A dated record of what was decided, moved and blocked is precisely the artifact an auditor or an assessor means by governance. The same twenty minutes of secretarial discipline that keeps the plan honest also feeds the next questionnaire without a scramble.

Run this way, the meeting gets shorter as the plan gets older, which is the correct direction. The first review is an hour of hard conversations. The fourth is twenty minutes, because the rows have owners who arrive knowing their answer, and the hard conversations happened upstream where they belong.

The two ways of appearing to act

Two moves reliably fill the space where a plan should be, and both deserve suspicion in proportion to their popularity.

The intensity-only target. Emissions per euro of revenue, per unit, per employee, falling nicely while absolute emissions rise with growth. Intensity is a legitimate steering metric, and we argued as much for industrial sites, but a company whose only target is intensity has committed to nothing about its actual emissions, and the atmosphere does arithmetic in absolutes. Publish both numbers and commit on the absolute one, even if the commitment is more modest.

The offset purchase. Buying credits equal to some share of the footprint and communicating an achieved neutrality. Whatever the quality of the credits, and quality varies enormously, the purchase changes nothing on your own inventory: your scopes 1, 2 and 3 are what they were. Contribution to projects outside your value chain can be a fine thing, done knowingly and named as contribution. As a substitute for the five-line table, it is the most expensive way available to not have a plan.

On science-based targets, the SBTi framework is useful discipline precisely because it forces absolute commitments on a timeline, and its simplified SME route removes much of the process cost for mid-sized companies. The reservation is sequencing: a company that has not yet run one ninety-day review cycle gains little from spending a year on target validation. Learn to make the number move first, then formalize the trajectory once you have evidence you can.

The diagnosis was the cheap part

Measurement is the essential first step, and a first step is all it is. The footprint told you where the emissions are. It did not negotiate the recycled-content clause, rewrite the travel policy or re-spec the concrete. Those acts have owners, dates and budgets, or they do not happen.

So end the measurement project the way treatment begins. Before the presentation meeting closes, put the five-line table on the screen and do not leave the room until every row has a name in it. The diagnosis is done; everything that matters now is scheduled.