You need to stock the shelves with products that your customers want to buy, but you don't need to create a Wal-mart size store right away. You are right to focus on one local market first. Figure out how many transactions you need to prove product/market fit and then work backwards. If you need 50 transactions, how many products or services or service providers do you need to get to 50? Be laser focused on who your early adopters are, and stock the shelves with what they are most likely to purchase first... good luck!

I've seen this model a lot lately (not a criticism, just a fact). The internet has certainly created plenty of opportunity to bring buyers and sellers of a particular product together and create an efficient marketplace. What you have, though, is a classic chicken-and-egg problem. You need buyers to attract sellers, and you need the sellers to attract buyers.

Without knowing the actual market / product, it's difficult to give much advice, but one thing that I've seen help in this scenario is to take your market, and further slice it into niches, and focus on one niche to begin with. This gives you a couple of advantages:

1. It allows you to get hyper-focused on who your customer is, making introductions / referrals easier. This is counter-intuitive, but the more narrow your focus, the more likely you are to get other people to understand what you are looking for and send prospects your way.

2. It allows you to get maximum bang-for-buck in advertising and marketing, because your audience is more narrow and easy to identify. You can also target specialized trade shows or other, similar marketplaces where that audience can be reached (again, I don't know any detail about your business, so I'm generalizing).

3. It allows you to refine the MVP with a very narrow audience before opening it up to a wider group after it's been proven. You may find that changes are needed in order to widen your audience / customer base, but that's OK - you've proven that the model works and is worth investing in further.

I would be glad to discuss your model in detail and give you some more specific examples of what I've seen work if you are interested in scheduling a call.

Starting in one very tight niche (e.g., as you suggest, one geographic area) is a good idea. But I’d also add another very important point:

Build one side of the marketplace first, by providing something of value (other than the marketplace itself, which won’t exist yet) to one side of the marketplace.

For example, if you want to build a marketplace matching taxis to users, build the network of taxis by providing a useful tool to taxi drivers to allow them to monitor their mileage / fares / etc. (which is what Hailo did). You can get hundreds or thousands of taxi drivers on your “network” (=using your tool) before you even mention the marketplace idea. Then when you are ready to launch to users, you already have your supply side in place.

Or provide a portfolio service for designers, so you get lots of designers hosting their work using your service. Then offer to match people who need designers with the designers on your network who are looking for more work.

This approach is covered in more detail in this Pandodaily post from 2012:

http://pando.com/2012/11/20/liquidity-hacking-how-to-build-a-two-sided-marketplace/