how much, how long, and how many people does a website like yelp cost to make?

[dudnaito]

any additional and nuanced information would be great. Thanks.

[Sarah G]

Well it's complicated. Are you talking about the engineering costs alone? Or all the business that needs to go into supporting a company like that? To run a startup, you don't just need to build a site, you need marketing & advertising, sales, web hosting, insurance, lawyers, accountants, etc. And you need managers to oversee your staff.

If you are talking about just the development, it also depends on several things. How much traffic are you expecting? How many users do you expect? How much data is involved? How well polished do you need the user experience to be? Do you need to interface with an ad exchange? Do you need mobile capabilities and how much? Are there discussion boards? Is there a whole separate interface for businesses? Do you integrate with social networks and the like?

You really need to think in terms of all the different features of your product, and then talk to a professional engineer to get details and estimate the effort involved in each one.

As an estimate, from a purely engineering perspective, you could probably do a basic Yelp knockoff with half a dozen good developers and launch after a year or so. Including payroll taxes, benefits, hardware, recruiting, etc, I'd ballpark something like $250 k / developer / year (depending where you live).

Throw in a manager and a product manager and we'll say $2 million to develop the software. Of course if you really want it to be as good as Yelp, the number will be *far* higher. If you're just looking for something very basic, it could be a bit lower.

However, without sitting down with an expert and hearing the idea in detail, you're not going to get a very good estimate. In general it's pretty normal for a major new offering to take 9-18 months to get to market from the time you start development. Anything public is going to require at least a few developers, so you're definitely looking at something on the million-dollar level for any serious software project.

(If it's cheaper, it's often not worth doing because you'll just get copied right away by one of the big guys like Google.)