How to read an EcoVadis scorecard without guessing

Two companies score 58. One has an environment problem, the other has a sustainable procurement problem, and their action plans have nothing in common.

That is the first thing a scorecard is for, and the overall number hides it. An EcoVadis assessment is not a grade on your company; it is an assessment of the management system you were able to document, across four themes: Environment, Labor and Human Rights, Ethics, and Sustainable Procurement. Read it as a diagnosis and it tells you what to do next. Read it as a grade and it tells you almost nothing.

Read the four themes, then the three dimensions

Ignore the headline number for a moment and look at the theme scores. The overall figure weights them according to your sector, size and country, which is why the same improvement is worth more to some companies than to others, and why comparing your total against a peer's total is close to meaningless.

Then look at how each theme is assessed, because every one of them is scored across three dimensions:

• Policies. Do you have a formal, dated, endorsed commitment on the topic?
• Actions. Are there measures actually in place, described concretely enough to be verifiable?
• Results. Do you report indicators, with figures, over time?

Most mid-sized companies score reasonably on policies, unevenly on actions, and poorly on results. Check that pattern against your own scorecard, because where it breaks is where your effort belongs.

Good to know: an activity that exists but is not documented is not scored. That is not a technicality. The assessment evaluates what a third party could verify, which is precisely what your customers are asking about when they request the scorecard.

Three questions per improvement area

The scorecard ranks improvement areas by priority, and that ranking already accounts for the weighting, which makes it the most useful part of the document and the part most often skimmed.

Take each item in order and ask three questions:

1. Does the thing exist? If not, this is a programme decision, not a documentation task.
2. Is it written down, dated and endorsed? If it exists but lives in someone's head or in a slide deck from two years ago, this is a documentation task, and it is the cheapest kind of point to recover.
3. Do we measure it? A policy with an action but no indicator caps what the theme can score, no matter how good the practice is.

The gap is rarely between what a company does and what good practice requires. It is between what a company does and what it can show.

What tends to be missing

Undated documents. A policy with no date and no signature cannot be placed in time, and an assessor cannot tell whether it is current.

Group documents used for a subsidiary. If the assessed entity is a subsidiary, a parent-company policy that never names it is weak evidence.

Actions without scope. "We train our employees on ethics" raises the questions of which employees, how often, and how many completed it. A training rate with a denominator answers all three at once.

Results with one data point. A single year shows that a measurement exists. Two or three years show a trajectory, which is what the results dimension is looking for.

A realistic order of work

Split the ranked improvement areas in two: the ones that need a decision, and the ones that need a document. Do the documents first, because they are fast and they free up attention for the rest. Then take the decisions, each with a date, and record them as commitments with owners.

So aim the next cycle at evidence rather than initiatives. The thing that most reliably moves a score between two assessments is not a new programme, it is turning practice you already have into dated, scoped, measured evidence, and adding the second year of data to indicators you already collect. That will not guarantee a medal, since ratings depend on the methodology in force and on how peers move, but it does stop you losing points for work you have already done.