http://www.xconomy.com/boston/2009/09/11/techstars-first-class-of-boston-startups-launched-at-microsoft-hosted-gala/
"Baby monitors are a $350 million market in the US"
Is this a wild-assed guess, or are these sorts of things tracked and tabulated somewhere? When looking to start a new business, where do hackers look to discover the size of their target markets?
http://www.census.gov/main/www/access.html, http://www.bls.gov/data/, http://www.sec.gov/edgar.shtml
In my experience the best way is to analyse a market in several different ways and get bounds on the market size. E.g. for baby monitors in the US, the market size must be less than the total revenue of all the companies that make them, and the units sold per annum must be less than the number of babies born per year in the US. You then may find one company that claims to have shipped over a million baby monitors in a year (e.g. inside their SEC filings or even marketing material), which is therefore a lower bound.
A combination of google, asking people (e.g. either industry experts or in your local shop ask the shop assistant roughly how many of those do they sell a day) and you should be able to get to an order of magnitude of an established market place.
The other thing is to consider Total Addressable Market (TAM) vs (SAM) Served Addressable Market. Basically, it is the difference between a complete market and the portion you would be able to serve. This is helpful to keep in mind if you are targeting a niche market. The SAM niche may be small, but if there is applicability to the TAM it could be viewed more favorably.
Here is a good link describing SAM vs. TAM:
http://windowmanager.blogspot.com/2005/05/wth-is-tam-sam-som...
Its a whole lot of voodoo and has little to do with anything that matters to your business, since for a market of any size you're going to be limited by how much of it is addressable anyhow.
I swear, the only reason people even bother checking is so they can present the "if we got 1% of $350 million..." argument that everyone says they know is useless.
You can make some rough estimates by figuring out what substitutes your product has, and then researching the combined revenues of all the products you'll be replacing. This also degenerates nicely: if you can't think of any substitutes, then it's fairly likely that you have a $0 market.
Does the market exist? Does the need exist?
For startups, these last two questions are most useful in my experience: trying to attack a large market because it is well established is not normally a recipe for success unless you have some way to take advantage of a new trend or a new technology.
I would also focus much more on what's required for you to reach break even operation in a niche or segment since failure to do so will kill you more quickly than many other things.
TAM is size of your company's market * price of 1 unit of your product.