 |
| David Cuykendall Click for Index Page |
A target is what we want to happen. A forecast is what we think will happen, whether we like what we see or not. And resource allocation is about trying to use our resources in the most optimal and efficient way.
The solution to this problem is dead simple: Separate the three purposes, which make it possible to optimize each one in much more tailored processes. This allows for different numbers, updated on different frequencies and time horizons in each of the three processes.
Conventional budgeting uses stale information to hierarchically cascade authority
In conventional budgeting, documents and reports course their way through organizational lines of authority. The paradigm is the same paradigm of standard compliance oriented accounting systems. Budgetary data is captured in source systems, scrubbed, funneled upward, and then posted to repositories where it is queried and summarized. It is all done on an annualized calendar basis, not an event basis. It is a very time consuming and arduous hierarchical process.
Conventional budgets are hierarchical control documents — they cascade authority on the basis of what is, in essence, upfront funding of a fixed performance contract.
The main purpose should not be centralized command-and-control, but rather a management model which is very well-suited to dealing with turbulence and rapid change, enabling action and reprioritization quickly to fend off threats or seize opportunities.
January-December is an artificial construct from a business point of view. For some it is too short, for others too long. Even when a business has seasonal rhythms, the winter season is cut in two because we pass “year-end”.
Further, on-demand allocation of resources, as opposed to upfront funding, minimizes costs because estimates, forecasts and trends are updated at the exact time they are needed. On-demand resource allocation provides a much bigger and more flexible decision authority and a much more dynamic rhythm.
Imagine a finance guy meeting a fisherman, asking him about the rhythm of his work. “Well,” the fisherman replies, “I am at sea for five months, and then I am home for five months.” “So what do you do then the rest of the year?” the finance guy wonders. Something is wrong, right? Absolutely, but maybe more in the head of the finance guy than in the working rhythm of the fisherman.
Our statutory accounting and our communication with external parties and the capital markets will of course still need to be calendar oriented, but our internal processes could still have more natural rhythms. We need to free ourselves from the artificial, annual "stop/start/stop/start". We need to give the opportunity to run our businesses more continuously, with update frequencies, time horizons and evaluation points driven by the natural business flow.
No annualized planning and budgeting
Examples of non-calendar driven boundary setting for costs and profits include “burn rate” guidance ("operate within this approximate activity level"), unit cost targets (“you can spend more if you produce more”), profit targets (“spend so that you maximize your bottom line”) or simply no target at all (“we’ll monitor cost targets and cost trends and intervene only if necessary").
The main principles the new and fully dynamic processes are as follows: No annualized planning and budgeting. Strategic objectives, KPIs, and targets and forecasts can be changed when deemed necessary by the business units themselves — event-driven changes, not calendar-driven.
Events can be external or internal, and the definition is simple: “Whatever is important for your business unit."
There should be simple change and cross-business unit coordination controls: Big changes should be approved one level above, smaller changes only informed about. Big or small, always inform other affected units if necessary.
Business units sort out between themselves what is big and what is small. Target and forecast horizons should reflect lead times, urgency, uncertainty and complexity in the different businesses.
This gives rise to the possibility to change targets at any point in time, perhaps because something happened and they lost their meaning. That could either mean "impossible to achieve" or "a piece of cake", or also "not relevant anymore."
A target should motivate and inspire. It is not a goal in itself, but one way of achieving the ultimate goal of the best possible performance, given the circumstances.
Targets set by business units themselves typically do this job much better than those coming as instructions from above. First, there must be full transparency. Second, you still need approval for big changes. And third, you don't need to change your target every time assumptions change. A relative target is much more robust and self-regulating in this respect. If conditions change, they typically do so for your peers as well.
Business units need to review their track-record of changing targets. If it always is about reducing ambition levels and never the opposite, that is an issue the business unit should reflect on.
Dynamic forecasting
Dynamic forecasting is different from what is often called rolling forecasting. A rolling forecast is done on a fixed frequency and on a fixed time horizon across a company, often quarterly and five quarters ahead.
In contrast, dynamic forecasts are updated when something happens, and as far ahead as relevant for each unit. For some that could be short, for others long. If a unit one level up needs a forecast with a longer time horizon, it is their responsibility to “fill the hole” with a “good enough” forecast. Why should business units be forced to look ten years ahead, because aggregated, this is a relevant time horizon for a particular company?
For many people, anything beyond three weeks can be quite foggy, while three years is very much on the short side for other enterprises. Also targets can have shorter or longer time horizons, from months to years, again driven by lead times and complexity.
Note that "dynamic" doesn’t necessarily mean more often. It means at the right time. For some it could actually mean less often. Performance against targets is reviewed monthly or quarterly in business review meetings. A log in the business unit’s role charter (which actually is an internally developed shared document) keeps an overview of changes made.
Why is once a year in January always the natural point for evaluating performance? Is it only because of the link to pay? Is that a good enough reason? Are there alternatives, again driven by more natural business milestones?
This planning work is not about budgets but about action planning and understanding the consequences of actions through unbiased forecasts of expected outcomes. The goal is a more of a “living” forecast where those needing forecast information can tap into the latest information available. This can be done both regularly and ad-hoc. This means taking steps of radically challenging the calendar rhythm, aiming for a fundamental break with the calendar year.
It means business units have more autonomy and flexibility, but also higher accountability for results.
