Every quarter when earnings-release time comes around, Tim Cook has been remarkably accurate with his forward-looking guidance for the next quarter. As he should. Accurately forecasting Apple’s financial performance for the quarter not yet complete is important for one big reason. It shows that the executive team knows their business.
In order to forecast you have to have your finger on the pulse of leading indicators and you have to have an accurate set of real-time metrics. But the goal isn’t just to give Wall Street guidance. This information allows Tim and his team to identify any negative trends that may start to form in the business. If there is a hiccup in material purchasing or manufacturing, the management team needs to address the issue as soon as possible to minimize any financial damage. Accurate guidance is a company’s way of proving to the financial world that they are actively minding the store.
So how does Tim do it? How can they accurately predict only 3 or 4 weeks into a quarter what things will be like months later? Unfortunately, there is no crystal ball involved. I’ve been forecasting for over twenty years at many Fortune 500 companies, and the drill is fairly similar at most places. Here are some major factors.
The “Revenue Recognition” Effect
Apple is further into the quarter than most people realize. You would think that since Q3 ended on June 25th that when Tim gives his guidance on July 26 that Apple is four weeks and three days into Q4 right? Wrong. Accounting doesn’t work that way.
Companies don’t get credit for their sales until they have invoiced their customer and the products have shipped. But accounting goes a step further and make the assumption that items shipped in the final days of the quarter probably won’t be received by the customer until the following quarter. So what do we do? We remove sales revenue for the final week of the quarter and throw them into the next quarter. So the last week of Q2 will be included in Q3. But Q2 isn’t short, because it has the last week of Q1. And so on.
This means that by the time that Tim gives an account to Wall Street, he has not only four weeks of sales but probably over five. That’s a pretty significant chunk of the quarter already history. Financial quarters always start and end sooner than the official dates due to the revenue recognition adjustments that the accounting group will do behind the scenes.
Institutional Sales and Just-In-Time Inventory Shipments
Nobody wants to carry excess inventory any more. That means that they are relying on the fact that they can order inventory and have it delivered on short notice. So how do Best Buy, Verizon, or AT&T accomplish this? They will hand over a twelve-week forecast of their expected purchase orders to their supplier so that they have no excuse not to have the inventory ready when they place an order. Or better yet, blanket purchase orders are set up to allow for automatic shipments.
So not only does Tim have five weeks of history, he has an order backlog from his biggest customers already in the system just waiting to get shipped. Plus, these large customers will have liaisons who will stay in constant contact to gauge any sales-trend shifts from the customer and communicate those back to the various departments within Apple that need to know.
Consumer Sales
The trickiest part for Apple is going to be forecasting all the individual customers who order directly from them. They will have some kind of “installed base” model to help them predict how many upgrades they’ll get from devices due for replacement. They’ll also rely a great deal on historical models of seasonality. And you’d be surprised how cyclical most sales are. But those historical models can’t help you quantify the impact of new market conditions or competitor marketing promotions. This is where the sales forecasters really earn their money.
When an army sniper demonstrates for his superiors that he can hit a bulls-eye from a mile away, the point isn’t to hit a piece of paper. The point is proving to your superiors that in a time of war you can get the job done. Similarly, when Tim and company can throw out guidance and then hit it square on the nose twelve weeks later there is a message being sent. A message that says “We know this business well. We have the right people and processes in place to monitor trends and if anything comes up, we’ll immediately take care of it.”
So how does Tim do it? How can they accurately predict only 3 or 4 weeks into a quarter what things will be like months later? Unfortunately, there is no crystal ball involved. I’ve been forecasting for over twenty years at many Fortune 500 companies, and the drill is fairly similar at most places. Here are some major factors.
The “Revenue Recognition” Effect
Apple is further into the quarter than most people realize. You would think that since Q3 ended on June 25th that when Tim gives his guidance on July 26 that Apple is four weeks and three days into Q4 right? Wrong. Accounting doesn’t work that way.
Companies don’t get credit for their sales until they have invoiced their customer and the products have shipped. But accounting goes a step further and make the assumption that items shipped in the final days of the quarter probably won’t be received by the customer until the following quarter. So what do we do? We remove sales revenue for the final week of the quarter and throw them into the next quarter. So the last week of Q2 will be included in Q3. But Q2 isn’t short, because it has the last week of Q1. And so on.
This means that by the time that Tim gives an account to Wall Street, he has not only four weeks of sales but probably over five. That’s a pretty significant chunk of the quarter already history. Financial quarters always start and end sooner than the official dates due to the revenue recognition adjustments that the accounting group will do behind the scenes.
Institutional Sales and Just-In-Time Inventory Shipments
Nobody wants to carry excess inventory any more. That means that they are relying on the fact that they can order inventory and have it delivered on short notice. So how do Best Buy, Verizon, or AT&T accomplish this? They will hand over a twelve-week forecast of their expected purchase orders to their supplier so that they have no excuse not to have the inventory ready when they place an order. Or better yet, blanket purchase orders are set up to allow for automatic shipments.
So not only does Tim have five weeks of history, he has an order backlog from his biggest customers already in the system just waiting to get shipped. Plus, these large customers will have liaisons who will stay in constant contact to gauge any sales-trend shifts from the customer and communicate those back to the various departments within Apple that need to know.
Consumer Sales
The trickiest part for Apple is going to be forecasting all the individual customers who order directly from them. They will have some kind of “installed base” model to help them predict how many upgrades they’ll get from devices due for replacement. They’ll also rely a great deal on historical models of seasonality. And you’d be surprised how cyclical most sales are. But those historical models can’t help you quantify the impact of new market conditions or competitor marketing promotions. This is where the sales forecasters really earn their money.
When an army sniper demonstrates for his superiors that he can hit a bulls-eye from a mile away, the point isn’t to hit a piece of paper. The point is proving to your superiors that in a time of war you can get the job done. Similarly, when Tim and company can throw out guidance and then hit it square on the nose twelve weeks later there is a message being sent. A message that says “We know this business well. We have the right people and processes in place to monitor trends and if anything comes up, we’ll immediately take care of it.”
RSS Feed