Reports don't steer. The steering rhythm does

The monthly report is ready on the eighth working day. Twelve pages, reconciled to the close, prior year – actual – budget on every line, notation following IBCS®, three KPIs red. It goes to the management team on Wednesday evening; the meeting is on Thursday. The twelve pages are leafed through, the head of sales explains order intake, the head of production the material ratio, the CFO liquidity. Everyone nods and the minutes record: "report discussed". A month later the same three KPIs are red, with the same explanations. The report was good. It steered nothing.

Why a monthly report alone steers nothing

The scene is the normal state in many SMEs that have put their reporting in order. The report is finished; the steering is not. Ossola-Haring, Schlageter and Schöning put it in one sentence: an automatically generated management report, a polished management cockpit or a glossy presentation full of facts and charts does not steer a company (Ossola-Haring et al. 2019, p. 96). Steering needs a process in which a variance becomes an action – and the action, a month later, becomes a checked effect.

Three things are missing, almost always together. First, the meeting: the report is one item among ten, without an agenda of its own. Second, the owner: the red KPI has someone who explains it, but nobody who must deliver something before the next meeting. Third, the response path: nobody has defined the variance at which action is taken and what happens then. So every number gets the same treatment – explained, not worked on.

Do the arithmetic. Twelve reports a year with three red KPIs each make 36 signals. If, as we counted in one review, four actions with a deadline come out of them, the reporting has a hit rate of about 10 %. The rest is number discussion – interpretation gets more attention than implementation, as Gunhild Posselt notes from her own management practice (Posselt 2017, p. 29). Weber and colleagues describe the same finding for large corporations: steering systems fail because they are driven by the instrument rather than by the question of how it should change managers' behaviour (Weber et al. 2017, p. VI). The instrument is the report. The behaviour happens in the meeting.

In the framework From Decision to KPI these are the steps Effect and Review – the two most often missing from reporting projects, because they have nothing to do with the tool. A project review that stops at the report therefore falls short; the question is always also: what happens the day after distribution?

The steering rhythm has three beats

The pattern that has proved itself in our engagements is unspectacular: a monthly beat for the management team, a weekly beat for operations, a quarterly beat for the KPI set itself. Every KPI has a cadence – "how often do you look at it?" belongs in every KPI profile – and every cadence has its meeting.

Monthly beat: from the close to the action list

The monthly beat is a chain with six links. Close → report on a fixed day → steering meeting with a fixed agenda → action list → implementation → effect check at the next meeting. The steering meeting is a meeting of its own, 60 to 90 minutes, not an item in the weekly stand-up. It deals only with variances above the threshold – in our work usually ±5 % against budget or forecast. Anything within it is not discussed; the target is the baseline, and the value only matters as a deviation from it.

Ossola-Haring et al. describe the traffic light for this precisely: green is passed over, amber is discussed – even when the variance is compensated at the aggregate level –, red triggers an action in the action list. And the owner may override the automatic colour, with a reason (Ossola-Haring et al. 2019, pp. 96–99). The override is not a loophole; it is where context enters the number. A red order intake explained by a large order already signed for next month is amber – and the owner stands behind that.

Agenda of the monthly steering meeting
Item	Input	Outcome	Duration
1. Action list from last month	Action list with status per action	Done, extended or dropped – with the effect on the KPI	10 min
2. Traffic-light overview	One page, all steering KPIs, traffic light and overrides	List of KPIs to be dealt with today (amber and red only)	5 min
3. Variances per KPI	Variance against budget/forecast, owner's hypothesis on the cause	Cause confirmed or rejected; decision: action, observe, or adjust forecast	30–45 min
4. New actions	Owner's proposal	Action with owner, deadline and target value in the action list	10 min
5. Forecast and outlook	Forecast changes from item 3	Updated forecast as the reference for next month	10 min

The agenda starts with last month's effect, not with the new numbers. Green does not come to the table; every red KPI leaves the meeting with an owner, a deadline and a target value, or with a reasoned decision to observe it.

For item 3 we use the six steps Posselt describes for dealing with variances: gather the facts and clarify the context – is the situation really as the KPI shows it? –, form hypotheses, validate them with the people involved, keep the scope of change small, work out actions towards the unchanged target, and reach consensus (Posselt 2017, pp. 129 f.). Two of these steps discipline the meeting in particular. The first prevents a reaction to a data error. The fifth prevents the most common outcome of number discussions: the target is adjusted downwards, and the light is green again.

Weekly beat: the hammer runs without the management team

Not every KPI belongs in the monthly beat. We distinguish three kinds of report: the glasses (dashboard, overview for the management team), the page (report, analysis for a unit head) and the hammer (working report, operational tool). The hammer has a weekly or daily beat: backlog, delivery reliability, receivables over 60 days, capacity next week. These numbers need no management meeting, but a fifteen-minute unit huddle with the same logic – threshold, owner, action – and an escalation rule: red three weeks running goes to the monthly meeting. That keeps the glasses to one page, and the management team does not discuss picking errors.

Quarterly beat: the test bench for the KPI set

Once a quarter it is not the business that goes on the test bench but the KPI set. The question per KPI: did it trigger a response in the last three months – an action, a forecast adjustment, a reasoned decision to observe? If not, and nobody can name a decision that depends on it: strike it or move it to the appendix. Eric Ries describes for start-ups a regular pivot-or-persevere meeting in which the metrics serve an explicit decision on whether to change strategy – and warns that without a fixed cadence and a predefined expectation the decision is postponed indefinitely (Ries 2011, ch. 7 f.). The quarterly test bench is the SME version: pivot or persevere not for the business model, but for every single KPI.

The report hangs on the close

A steering meeting with a fixed agenda only works if the report arrives on a fixed day – and that only happens if the close follows a schedule. That is why we start every reporting operating model at the close, not at the meeting. The schedule we work with uses relative days instead of calendar dates: day −30 to day +120, per task an owner, a deputy, the documents and the system transaction, plus a matrix per month showing whether the task is due. Monthly, quarterly and annual tasks sit in the same list.

A few marks on it matter for the rhythm. Day +1: close system periods, switch report parameters. Day +8: cost accounting settled, variances posted – from here on the result figures are reliable. Day +10: report distributed. Day +12: steering meeting. Day +15: forecast updated. Send the report on day +6 because that is when the meeting is, and the meeting will negotiate the numbers, because they keep changing until day +8. Reconcilability comes before beauty – and before speed. The gap between close and meeting can be shortened, but only by shortening the schedule, not by pulling the report earlier.

Measuring the effect: three numbers about the reporting itself

Whether the rhythm is taking hold can be checked with three numbers we build into every steering operation. First: actions with owner and deadline per meeting – the hit rate from the start of this article. Second: share of actions completed by their deadline. Third: share of steering KPIs that triggered a response at least once a year – the input for the quarterly test bench.

Fictitious example: in the prior year, without an action list, fewer than half of the agreed actions were implemented on time – and the figure did not move. With a fixed agenda whose first item is the action list, the share rises above the 80 % target within six months.

The curve is fictitious; the pattern is not. The share rises not because the actions get better, but because they are back on the table at the start of the next meeting. That is the whole mechanism: whoever expects to be asked about their action in four weeks implements it or says in time why not.

From practice

At an industrial company in building technology with SAP and a separate finance system, we had unified the management reporting: one page per unit, prior year – actual – budget, reconciled to the close. The report was finished, and the management meeting still remained a number discussion – every unit head explained their page, and there were hardly any decisions. I then proposed rebuilding the meeting itself: a traffic-light overview as the first page, only amber and red on the agenda, and an action list walked through at the start of every meeting. For two months it felt unfamiliar, because suddenly there were deadlines in the minutes. From the third month the meeting got shorter, and for the first time KPIs left the red zone because somebody had done something – not because the month was better.

Posselt notes that variances are impulses for change and, as such, normal – no company hits its plan figures 100 % (Posselt 2017, p. 129). That sentence belongs at the start of every steering meeting because it frees the meeting from justification: the variance is not the problem, the variance without a response is. Punish variances and you get embellished numbers; turn them into actions and you get steering.

What you can do tomorrow

Take the minutes of the last three management meetings and count the actions with an owner and a deadline that came out of the monthly report. Fewer than one per red KPI means the rhythm is missing, not the report. Set the five items from the table as the agenda for the next meeting, with the action list first, even if it is empty the first time.

If you would rather not build the rhythm alone: in our steering operation we run the rhythm with you – monthly reporting on a fixed day, steering meeting with agenda, traffic light and action list, quarterly test bench for the KPI set. An initial conversation clarifies where your closing schedule stands and which beat suits your business.

Sources

Ossola-Haring, C., Schlageter, A., Schöning, S. (2019): 11 Irrtümer über Kennzahlen. Mit den richtigen Erkenntnissen führen. Springer Gabler.
Posselt, G. (2017): Renditeorientierte Führungsstrategien. Kennzahlen motivierend einsetzen. Springer Gabler.
Ries, E. (2011): The Lean Startup. How Today's Entrepreneurs Use Continuous Innovation to Create Radically Successful Businesses. Crown Business.
Weber, J., Bramsemann, U., Heineke, C., Hirsch, B. (2017): Wertorientierte Unternehmenssteuerung. Konzepte, Implementierung, Praxis-Statement. 2nd edition, Springer Gabler.