Every quarter, investors pour over actual results and react to a company’s guidance. Before it appears in the press release, I have to work out what is should be. I often see retail investors wonder where the numbers come from and if companies sandbag or otherwise fool around with guidance. So I decided to write a post on the process I have followed to first support the process when I was a Controller and then to help decide as a CFO.
Helping to decide is important to remember. It is easy to get a little ahead of yourself as CFO during the earnings and forecasting process. Take a deep breath and remember that you are part of a senior management team. Remember that there are two types of forecasts, lucky and wrong, and that the longer period of time being forecasted the higher chance you have of being wrong.
I start my outlook process by asking questions and listening. Of course, I keep up with the business as best I can during the quarter, but the reporting process is where you get to ask questions and explain. Take advantage of that and start to form an idea of the opportunities and threats that are coming up.
I am going to assume that you have a reasonable forecasting system for revenue and cost that you are able to rely on to make management decisions. How the actual forecasting is done is not really relevant (Gaussian Cupola or whatever wizardry is used), you just need something that has proven it works in the past. Units, ASP, expected costs, any expected deviations from your normal SG&A spending. Some form of view on what effect foreign currency is likely to have.
You should use the information from your forecasting system to build a draft forecasted financial statement. This is the raw and preliminary numbers that I rely on to start the judgment process of the outlook I eventually publish. Using that first set of numbers, I start to work through a standard checklist of accuracy items (not judgment yet):
1) Revenue – usually the internal forecast is good at predicting when the goods will ship. What is not considered is when you can recognize it. Double check that. Look at the terms being used and any carry-over from the prior period. Be careful with percentage of completion and double check the reasonableness of the calculation. Make sure that items that require contract analysis have been reviewed and that the terms allow revenue recognition as hoped for. Make sure any inter-company sales are properly treated.
2) COGS – usually you get the latest cost numbers from the factory, but that probably is not what will show up in GAAP. Very often their numbers do not include warranty reserves, equity compensation, inventory reserves and potential equipment write-offs (which may need to go into COGS). If costs are rising or falling there will be a lag as production runs through inventory so make sure that is accounted for. Any inter-company profit in inventory that needed to be eliminated? Inter-company items are often overlooked, so check that again.
3) SG&A – talk through the numbers and see if there are unusual things happening this quarter and make sure the numbers are there. If there is a risk of a large A/R write-off or a large fee that is being paid that quarter that needs to be expensed, make sure it is included. Think about bonus accrual catch-ups, if you are beating or missing bonus targets is there a risk that an adjustment to the accrual will be triggered this quarter. Are there large asset write downs, big A/R wrote-offs that could happen or other such one time items? If so, are they in the forecast (of course, if you are so sure that it will happen, ask why is it not recorded in the about to be reported quarter). These type of forecasts are often just run-rate extrapolations, so make sure any effect of adding or selling a business recently was thought through and recorded.
4) Interest and other charges – normally this should be about the same as prior quarters. Double check to make sure any changes or expected changes are reflected in the numbers. Usually companies do not forecast specific foreign exchange swings but if there already has been a large one or a large one is likely you probably should be modeling here.
5) Taxes – typically this is just a forecasted percentage, but work through the logic of the forecast. Did the country mix change since the prior quarter or since the percentage was last set? If so, you may need a different rate and should let people know to change their rate.
6) EPS – double check the diluted EPS calculation. If you recently added convertible bonds, check to make sure the EPS is on an “if converted” basis. Most people just assume that there only is dilution if the bond is in the money and that is not the way the share count is determined for convertible bonds. If you are close to a trigger point for a large tranche of options to go in the money, consider reminding the reader of that. If they have been outstanding long enough to be vested and around for several periods there may be enough added in to swing the EPS calculation.
That is not an exhaustive list. Companies that sell software and other service type contracts need a robust revenue recognition review process for both actuals and forecasts. The list above is all for accuracy before discussing with the greater team. It is all fine and nice to be a strategic CFO but if the base numbers are all wrong then no one will listen to your business ideas.
If you do find errors during the forecast review, go back and fix what did not work in your process. Double check the actuals as the same people usually work on actual and forecast and if the forecast has an error, the actuals might as well. Reporting is a process and you need process controls and feedback into the process.
I am sure that some people started reading this blog and we’re hoping that I would talk about what system to use, what IT tool I recommend relying on. I am always concerned about over reliance on tools and not enough review and finance judgment being applied. A good tool poorly used just gets you the wrong answer faster. There are quite a few good tools out there and I generally recommend that you use one that works well with your existing accounting and consolidation system. Most of my experience is using the Hyperion suite of tools (Oracle product now but used to be independent), but I have used others as well. Even simple excel spreadsheets can work depending on your business. I do think that if the Excel spreadsheets get too big, the risk of formula errors also grows too much.
So now you have a forecast (I tend to have an income statement and balance sheet ready with a cash forecast from another process as well). You now move to the validation and judgment phase. Share the forecast with a small number of senior executives and start the discussion with them. Do not over rely on email. Call them up and talk through the forecast.
Go through each section with them and make sure they understand what assumptions are being used. Then listen. They know their area of the business better than you. The COO probably just heard that there is a part shortage and production will not be able to hit the cost target or the volume target. The SBU business head might have just come from a customer meeting as has good news. Take notes and ask clarifying questions but now is not the time to be pushing back or disagreeing. If you go into target mode and start arguing why results need to be better, you probably will shut down conversation and miss something.
One of the people you should be talking to now is your boss. The CEO will have even more insights into business conditions and the two of you will be making the public commitment. Review the different options from the management team, weighing factors like personalities, some people are too conservative and some are too optimistic. Reach a conclusion about what you will commit to in the quarter or year to come or whatever period the public outlook will be. Make sure the targets needed to hit the goals are rolled out to the people that need to execute.
One of the last things you need to decide on is the range of expected results you will disclose. You need to pick a range that is typical for the metrics you will be disclosing. I get asked very often if I set the range lower so we can easily achieve it and then “beat” the numbers. In my whole career I have not followed that strategy. It becomes very obvious over time if you do that consistently and analysts will start setting their own much more aggressive expectations. You are essentially lying to your investors about your real expectations and trust in management is often a key metric for investors. You also are advertising less than what you can do and if your competition does not under call as well, you will suffer in comparison.
Never set expectations you know you cannot meet or that would take perfect execution, but do not be afraid to set your own internal goals as the public expectation. If you miss or beat them, know why and explain it. If you set reasonable expectations that a good effort should have achieved and you miss, do not try and hide the fact that you missed. Be honest and outline what will happen in the future to avoid surprises and underperformance. If your internal process to arrive at the number was bad, fix that process. If you set expectations that are way too high, you may get a short term bump in share price but you will be reporting actuals in 3 more months and you will lose in trust much more than what you gained in the short term. Be confident, but not overly rosy or arrogant.
Pay attention to the text explaining the numbers to ensure that there is no disconnect. If you expect conditions to be tough, make sure the text reflects that. If you are setting another record, acknowledge it.
In the end, there is no magic formula. You need to do your best and make your best judgment. Make sure your process is robust and repeatable. Be inclusive of your finance team and the senior management team. Don’t sweat it too much, a miss or beat is news for maybe ½ a day if it is a slow news day.