background image
generally vary according to several factors including industry conditions, seasonal demand, competitor actions, chan-
nel mix and overall availability of product. Generally, total sales incentive and marketing programs range from 7% to
11% of gross revenues per quarter. For 2013, sales incentive and marketing programs were 8% of gross revenues.
Changes in future customer demand and market conditions may require us to adjust our incentive programs as a per-
centage of gross revenue from the current range. Adjustments to revenues due to changes in accruals for these pro-
grams related to revenues reported in prior periods have averaged 0.5% of quarterly gross revenue since the first
quarter of fiscal 2011. Customer sales incentive and marketing programs are recorded as a reduction of revenue.
We record an allowance for doubtful accounts by analyzing specific customer accounts and assessing the risk of
loss based on insolvency, disputes or other collection issues. In addition, we routinely analyze the different receivable
aging categories and establish reserves based on a combination of past due receivables and expected future losses based
primarily on our historical levels of bad debt losses. If the financial condition of a significant customer deteriorates
resulting in its inability to pay its accounts when due, or if our overall loss history changes significantly, an adjust-
ment in our allowance for doubtful accounts would be required, which could materially affect operating results.
We establish provisions against revenue and cost of revenue for sales returns in the same period that the related
revenue is recognized. We base these provisions on existing product return notifications. If actual sales returns exceed
expectations, an increase in the sales return accrual would be required, which could materially affect operating results.
We record an accrual for estimated warranty costs when revenue is recognized. We generally warrant our prod-
ucts for a period of one to five years. Our warranty provision considers estimated product failure rates and trends,
estimated replacement costs, estimated repair costs which include scrap costs, and estimated costs for customer com-
pensatory claims related to product quality issues, if any. We use a statistical warranty tracking model to help prepare
our estimates and assist us in exercising judgment in determining the underlying estimates. Our statistical tracking
model captures specific detail on hard drive reliability, such as factory test data, historical field return rates, and costs
to repair by product type. Our judgment is subject to a greater degree of subjectivity with respect to newly introduced
products because of limited field experience with those products upon which to base our warranty estimates. We
review our warranty accrual quarterly for products shipped in prior periods and which are still under warranty. Any
changes in the estimates underlying the accrual may result in adjustments that impact current period gross profit and
income. Such changes are generally a result of differences between forecasted and actual return rate experience and
costs to repair. If actual product return trends, costs to repair returned products or costs of customer compensatory
claims differ significantly from our estimates, our future results of operations could be materially affected. For a
summary of historical changes in estimates related to pre-existing warranty provisions, refer to Part II, Item 8, Note 4
in the Notes to the Consolidated Financial Statements included in this Annual Report on Form 10-K.
We value inventories at the lower of cost (first-in, first-out and weighted average methods) or net realizable val-
ue. We use the first-in, first-out ("FIFO") method to value the cost of the majority of our inventories, while we use the
weighted-average method to value precious metal inventories. Weighted-average cost is calculated based upon the
cost of precious metals at the time they are received by us. We have determined that it is not practicable to assign
specific costs to individual units of precious metals and, as such, we relieve our precious metals inventory based on the
weighted-average cost of the inventory at the time the inventory is used in production. The weighted average method
of valuing precious metals does not materially differ from a FIFO method. We record inventory write-downs for the
valuation of inventory at the lower of cost or net realizable value by analyzing market conditions and estimates of
future sales prices as compared to inventory costs and inventory balances.
We evaluate inventory balances for excess quantities and obsolescence on a regular basis by analyzing estimated
demand, inventory on hand, sales levels and other information, and reduce inventory balances to net realizable value
for excess and obsolete inventory based on this analysis. Unanticipated changes in technology or customer demand
could result in a decrease in demand for one or more of our products, which may require a write down of inventory
that could materially affect operating results.