Our policy is to book and report an order when a binding contractual agreement has been concluded with a customer covering, at a minimum, the price and scope of products or services to be supplied, the delivery schedule and the payment terms. The reported value of an order corresponds to the undiscounted value of revenues that we expect to recognize following delivery of the goods or services subject to the order, less any trade discounts and excluding any value added or sales tax. The value of orders received during a given period of time represents the sum of the value of all orders received during the period, adjusted to reflect the aggregate value of any changes to the value of orders received during the period and orders existing at the beginning of the period. These adjustments, which may in the aggregate increase or decrease the orders reported during the period, may include changes in the estimated order price up to the date of contractual performance, changes in the scope of products or services ordered and cancellations of orders.
The undiscounted value of revenues we expect to generate from our orders at any point in time is represented by our order backlog. Approximately 8.5 percent of the value of total orders we recorded in 2017 were “large orders”, which we define as orders from third parties involving a value of at least $15 million for products or services. Approximately 61 percent of the total value of large orders in 2017 were recorded in our Power Grids division and approximately 22 percent in our Industrial Automation division. The other divisions accounted for the remainder of the total large orders recorded during 2017. The remaining portion of total orders recorded in 2017 was “base orders”, which we define as orders from third parties with a value of less than $15 million for products or services.
The level of orders fluctuates from year to year. Portions of our business involve orders for long-term projects that can take months or years to complete and many large orders result in revenues in periods after the order is booked. Consequently, the level of large orders and orders generally cannot be used to accurately predict future revenues or operating performance. Orders that have been placed can be cancelled, delayed or modified by the customer. These actions can reduce or delay any future revenues from the order or may result in the elimination of the order.