Know your needs - "The decision is not wide-field versus confocal," says Watkins. "It's what you're trying to get out of the system." Figuring out the options can easily take a year. Talk to others who have set up similar experiments, and try out as many systems as possible. Or, suggests Maddox, reach out to live-cell microscopists who can translate your experiment into possible solutions.
Simplest first - "Start trying to image your sample with the easiest means possible" says Maddox. Before assuming you need a confocal, see what you can get out of a wide-field setup.
Don't bite off more than you plan to chew - A common mistake among imaging newcomers is to stock up on functionality your experiments don't require, says Sean Megason, who images developing zebrafish at the California Institute of Technology in Pasadena. Building too much into your system can make it less effective and more cumbersome. "Have a specific set of questions in mind that you plan to ask in the relatively near future," he suggests. "People end up paying twice as much and not using that extra ability."
Plan for problems - Having access to experts who can help when problems arise is just as important as having the right microscope. Most users we spoke with said product support varied widely between companies and was a primary consideration in the system they purchased. But support doesn't come free: For most confocal and multiphoton systems, you'll need to buy an annual service contract, usually costing between $20,000 and $40,000 per year.
J. Waters, "Live-cell fluorescence imaging," Methods Cell Biol, 81:115-40, 2007.