Readers of Consumer Reports and ConsumerReports.org have plenty to say—last year, 127,887 of our readers called, wrote, or e-mailed our customer-relations department with comments and questions about the thousands of products we test each year. In the home-and-garden area, they wondered about refrigerators that don't keep food cold, dishwashers that drown out conversations, and funny-looking twisted lightbulbs that claim to save energy.

The occasional rhetorical question arrives in our inboxes, as in "What the heck were you thinking?" (Actually, we get that genre of query pretty often, some of them not suitable to print here.)

Here we address some of the common inquiries we get about how we test products for the home. If you've got a question about a home-related product, send it to us at home@cro.consumer.org.

How do you pick the models you test?
We try to test models that represent the spectrum of products in a given market. Our analysts seek out products with new features and technological advances and a wide range of prices. After they analyze market share, marketing strategy, and advertising and promotional materials, they contact manufacturers to determine whether items will be available for at least three months after a report is published. The analysts then recommend a list of models that managers in our technical and editorial divisions review.

During the next step, staff shoppers buy the products at retail outlets in the throughout the Northeast—our offices are in the New York City suburbs—or online, never revealing that the purchases are for Consumer Reports. (We want to ensure that we test the same products you'll buy.) When we need to buy best-selling regional brands, we use shoppers around the country. Most significant, and unlike most other publications, we buy everything we test.

In rare instances, when a product isn't in stores yet, we buy it from the manufacturer, revealing this in our report; we'll subsequently test a version that we buy at retail and report on those findings.

How long does it take from when you decide to test a product to when you publish the results in a report?
Typically it takes about 9 to 12 months from the time we contact manufacturers for product information until a report appears in the magazine and/or online, a time frame that includes testing.

Note that for some fast-changing products like GPS units we can buy and test units and publish testing information in as few as two weeks. For other products that don't change as fast or for seasonal items (say chain saws), testing can take three to four months.

There are exceptions. Full testing of interior paints, for example, takes six months, but full testing of exterior paint requires up to three years.

How do you test?
Our experts develop tests that re-create the experience you'll have with the product. They also consider industry standards for testing a particular product; note that those tests usually gauge only a minimum level of performance while our tests aim to find the highest-performing products. We develop tests for those products that lack industry standards for ease of use.

In some cases, an industry models its tests on ours. For example, we developed an emissions test for vacuums that determines how much dirt and dust blows into the air when a model is running. The industry then devised its own test that’s based on ours, and now that test is the industry standard.

What do you do if a product malfunctions or breaks during the test?
When either happens we buy two more of the same product. If the new versions do not exhibit the same problem and we suspect the original problem was a quality-control issue, we base the results on the models that performed correctly; we chalk up the problem to an isolated issue. If either (or both) of the new samples exhibits the same problems, we make a judgment on whether it is a flaw in quality control or design and factor this into our Ratings.

A product-design flaw means that most consumers will experience problems with this item, while quality-control issues—materials, assembly, packaging, shipping—should not affect all of the products.