When doing a forecast each month, again and again, it can easily become a check-the-box exercise. But really it should not become one as this process should one of the key tools to drive an organization into a specific direction.
If a forecast it is made centrally without any input or cascading down of these forecast goals it will potentially loose its power and effect. In order for a forecast to become as close to reality it needs to be accepted by others as their personal goal. If they do not achieve the forecast there should be a feedback mechanism, a lessons learned or something more drastic if truly systemic. Why else do organizations spend man-hours after man-hours doing this exercise…
Below you find some background and several drivers that will help improve your forecast accuracy. As you can see it requires the entire supply chain to be involved in getting to a reasonably accurate number. To avoid loosing one-self into the detail, always sense-check your reasoning and assumptions to the big picture:
Why is forecast accuracy important?
- Forecast accuracy means you understand your business
- Understanding your business means you are reliable in your prediction
- Being reliable is what shareholders want: No surprises.
- Good forecasts allow the entire supply chain (and its scarce resources) to be managed more effectively
- Decisions are made of off forecast, hence the data needs to be as accurate as possible
- Start by sense checking your data, by using metrics, trend lines, graphs and ratio’s. “Does this increase in support #cost as a % of sales make sense?”, “Does our Sales in Q4 normally not increase significantly due to Christmas shopping?”
- Be aware of what is happening in your business and have the assumptions laid out. Talk to your business leaders, your sales people, and find out what is going on in your world!
- Understand the “why” behind what really happened. What can you learn from it. Why did we not know about this event when we built the forecast? Could we have known?
- Try to explain the operational facts, rather than #accounting terms. Using simple words will help your stakeholders understand what happened. Instead of saying “Bad Debt reserve adjustment”, say “Collection results from customers X, Y and Z have been delayed due late delivery”
- Get the detail behind the story line so you can articulate the results to others. Be specific, it will make you more convincing.
- Consider holding off changing forecast in later periods of the year to avoid sandbagging. People like to bring bad news first and not focus on the good news. Be very much aware of this dynamic, once you change the forecast this will become their new goal and they may give up on good news until the very end of the year.
- Forecast accuracy is great for ensuring accountability, but have a yearly plan to drive real year-end performance discussions. The plan is what you sign up for, not a monthly or quarterly forecast.
- Define management actions associated with the forecast and have follow up calls to ensure focus is being maintained.
- Communicate the new forecast and actions clearly and concisely
- Think like a shareholder…
- Help drive the business forward, think pro-actively and bring others together
- Have the worst performing units explain their variances to their peers or the leadership. Obviously this only works if the items that drove the forecast miss actually were controllable by the unit.
DRIVERS OF FORECAST ACCURACY
- What was the root cause of the forecast/plan variance?
- Which customers were driving the variances (requires you to know the baseline at customer detail)
- What happened to your backlog (open orders)
- What happened to your past dues (orders that were due to ship but didn’t)
- What is the root cause of your past dues (operational, credit, logistics, etc.)
- How much of the forecast miss is due to timing (e.g. we did not ship in June but will do so in July)
- What are the implications for future periods
- What actions can we take to recover sales (promotions, pricing actions, free products, discounts, etc.)
To know the following your planning process must have sufficient detail on your forecast / plan assumptions.
- What happened to our
- Product mix (e.g. sold more high margin products than low-margin products)
- Regional mix
- Plant mix
- Customer mix
- Channel mix
- Price realization (measure did we get price by comparing your prices from one period to another – typically harder to measure)
- Did our costs flex relative to the volume change as we would have expected?
- Did we adjust staffing levels appropriately to match the volume (e.g. through temporary work force or using 2nd / 3rd shifts)?
- Did we properly manage overtime?
- How has the tooling and supplies managed? Why did we spend more?
- What was our part mix and did we require more frequent change overs than anticipated in the forecast?
- Did we have any capacity issues or idle machinery driving the forecast variance?
- What portion of the variance was attributable to controllable vs non-controlable cost
- Was headcount the driver of the variance?
- Did logistics costs or other variable support costs flex with volume changes
- What activity centers and cost categories drive the variance?
- Where there costs recorded that should have been recorded in prior periods and what is the root cause?
- Were there any additional investments driving depreciation up?
- Did inventory move in line with volume realized and past due movements?
- Did we adjust appropriately for safety stock levels
- Did we have more or less Setup time?
- What happened to scrap and obsolete stock and scrap sales?
- Where there any scheduling issues
- Plant layout changes affecting the plan or forecast?
- Machine downtime?
- Where there any order entry issues?
- Was #cash applied properly?
- Did customers miss a payment date due to e.g. holiday’s?
- Is the customer paying late and yet taking discounts?
- Was there proactive customer follow up on frequent late payers?
- Were any customers put on hold not allowing us to ship and missing the forecast?
- Why are we paying earlier or later?
- Where there any invoices that were not recorded in the right period?
- Where there any invoices on hold in previous periods and released in the next driving a variance?
- Did we take the right purchase discounts or change our payment terms?
- Where there any circumstances that would impact the number of payment cycles in the period compared to expectation?
- Did we have large unanticipated capital investment invoices driving payables up or down?
Communicating the variance story:
Ok, so now we know what the issue was and why we did not achieve our forecast. We need to synthesize our message and tailor it to the right audience. Here are some questions you can ask yourself:
- What are the key message points I want to talk to? (materiality, relative importance, etc.)
- Looking back: what went well, what did not go well
- Outlook: what opportunities and risk are there to the plan/forecast
- Understand the objective of your message and your audience. Tailor your message depending on your audience.
- Comments should be fact-based, try to avoid emotional terms (unless you feel it is appropriate)
- Be concise in your presentation (e.g. use bullets) and be visual in your report out
- Quality checked and sense checked on argumentation of variance commentary
For those overseeing the forecasting process, they should remember that by definition a forecast will be wrong, but its is limiting the surprise factor that matters. Communicate the new forecast goals clearly and follow up regularly with the key stakeholders.
Remember that the only real tangible product the #finance function is our reports. We therefore should be committed to always deliver accurate and to-the-point actionable information to our stakeholders. These stakeholders will drive action with this data.
Here is a nice article from Analytics-magazine.org that gives you some more good insights in why forecasts are never right.
Good luck in managing your #forecast and planning process!
Write a story on Kingcashcow? Mail us at firstname.lastname@example.org