Focus on the more difficult side of the marketplace.

For instance, if you think it'll be easier to get suppliers, then focus first on getting buyers - always be working on your toughest problem (aka your biggest risk).

You'll find some great blogging on Marketplace and Platform topics here http://platformed.info (read the ebook too!)

Demand. You can always get supply once you have customers.

Might I suggest that you might have better success FINDING (versus "creating") a market.

And in that case you are discovering / uncovering a demand and then filling it. Just be sure that your business model is solid - because you can go broke filling (or trying to fill) a demand without it.

If you have massive resources (i.e. cash) and a protected proprietary solution and are skilled (or have a partner that's skilled) at creating markets - then go for it!

If, however, you are a typical start-up with limited resources and perhaps less experienced in growth-hacking / branding, then perhaps heed the advice here on clarity.

In either case... Best of luck to you!

It depends

When we started Wooshii we did very little to build supply. It was all about demand.

Our theory was "Build enough demand and finding supply is way easier."
It meant we didn't have to engage our community in other, (most likely distracting ways)

You can always deal with demand in an unscalable way early on. FInd some trusted suppliers to pick up the majority of the demand.

It worked out for us as we are service led. The last thing we wanted was a massive video community and no work for them

Another way to have done that might have been to create a sort of Single Player equivalent. Choose the side that has the most pain and help ease it with some other tool.

Of course with product or inventory type marketplaces supply may be the way forward - building an inventory of scarce collectables ( as eBay did) then allows you to start itching your product to the demand side

Number of posts

http://thenextweb.com/entrepreneur/2013/12/08/8-strategic-hacks-building-marketplace-liquidity/

http://techcrunch.com/2012/08/19/how-to-structure-a-marketplace/

http://karlinvc.com/blog/how-to-build-a-defensible-two-sided-marketplace/

I'm the co-founder of www.crated.com a marketplace for photographers and digital artists.

I'm going to say SUPPLY. Simple reason if your buyers land on your site and don't find anything you're done. They won't come back. Also the supply side users can be instrumental in helping promote and drive traffic to your site in the early days by putting them to work.

I used Clarity to speak with Micha Kaufman founder of Fiverr and he's a big reason for us shifting our strategy to building the supply side well in advance of launching our site which is currently in beta. He strongly suggested building SUPPLY.

Really surprised to see Dan Martel saying DEMAND since I know he worked hard to get the first 1000 experts on board before launching. Sorry to call you out Dan :-)

Supply! Supply! Supply!

IMHO.

Adrian :-)
Co-founder
CanvasPop / DNA11 / CRATED

Validate that there is a problem you are solving for both sides of the marketplace. Learn which one of the two sides is more 'hungry' for your solution. Then focus on the one side that is less hungry. It sounds counter intuitive but it's really quite simple. It'll always be easy to get the side that is 'starving' to use your product /service but not true for the opposite.
Example: UBER. It's pretty obvious that any taxi driver who'll be promised higher fulfillment rates (customers) consistently over time will want to use UBER. Their survival (pay check) depends on it! What's not obvious is that folks would pay a premium to get a cab under 5 minutes. Uber had to build the initial fleet of cars in one city but then the focus turned to validating that there was demand for the can-ordering service before scaling supply further.

Obviously in a new market, you need to be ready when you get your first order. So supply is critical. However too many companies shy away from demand building because it's not what they know how to do. They're not marketers and selling scares them. So they gravitate to what they know - the product - which leads them naturally towards the supply side. That's a waste of time when there is no market.

Demand, hands down! Here is how I would do it:

1. Pick a niche service specialty. If you're just starting out, determine which service speciality can get you the most early traction through secondary research. Make sure your basic premise (macro-level thesis) is solid.
2. Create your customer persona
3. Validate that by speaking with 30 prospects (or a statistically significant sample size).
4. Pivot (if need be)
5. Build your supply and create liquidity
6. Generate transactions and ...
I can go on about this, but I'm sure it will be never ending. Happy to talk more about this, if it would help.

Build the universe you want to live in.... No one cares about supply until they do. So focus on that "do". Build demand. The rest will follow.

You'll first need to define what side of YOUR marketplace is demand and which side is supply. For example, for DonorsChoose.org, I define the teacher need as "demand" and our donors' donations as "supply". In our case, we needed to have enough demand (teacher projects) before we started advertising such need to our donors. However, once we had enough demand (i.e. choice for our donors) it then became difficult to find the "supply" to meet that need, so we had to re-assign our resources from finding more teachers to finding more donors. You'll find that you'll always have an imbalance in your marketplace and you'll need to have enough demand and supply for the marketplace to thrive. You'll need to shift your resources from "demand outreach" to "supply outreach" (and viceversa) based on how your marketplace is behaving.

From a lean startup approach, try to create the side of the marketplace with which you have the most questions or doubts.

Tom Eisenmann at Harvard Business School has written a few papers on this subject. Try this one for starters. http://hbr.org/2006/10/strategies-for-two-sided-markets/ar/1

We've done a fair amount of this at Rev, building both supply and demand first for remote services.

Find the big gorilla problem(s) that need to be solved. From there one can assess the size of the demand. Without demand for a specific solution to a specific problem, we have nothing. When we have demand exceeding supply, we have the ingredients of a potentially profitable business.

Focus 1st on the part of the marketplace that is not automated to self-serve itself. You will learn from manual implementation and hence develop ideas to automate and scale quickly

Price is dependent on the interaction between demand and supply components of a market. Demand and supply represent the willingness of consumers and producers to engage in buying and selling. An exchange of a product takes place when buyers and sellers can agree upon a price.
At any price below P, the quantity demanded is greater than the quantity supplied. In such a situation, consumers would clamour for a product that producers would not be willing to supply; a shortage would exist. In this event, consumers would choose to pay a higher price in order to get the product they want, while producers would be encouraged by a higher price to bring more of the product onto the market. The result is a rise in price, to P, where supply and demand are in balance. Similarly, if a price above P were chosen arbitrarily, the market would be in surplus with too much supply relative to demand. If that were to happen, producers would be willing to take a lower price in order to sell, and consumers would be induced by lower prices to increase their purchases. Only when the price falls would balance be restored.
A market price is not necessarily a fair price, it is merely an outcome. It does not guarantee total satisfaction on the part of buyer and seller. Typically, some assumptions about the behaviour of buyers and sellers are made, which add a sense of reason to a market price. For example, buyers are expected to be self-interested and, although they may not have perfect knowledge, at least they will try to look out for their own interests. Meanwhile, sellers are profit maximisers. This assumption limits their willingness to sell to within a price range, high to low, where they can stay in business.
Let us take a few examples to explain my point:
Example 1. When a bumper crop develops, supply shifts outward and downward, more product is available over the full range of prices. With no immediate change in consumers' willingness to buy crops, there is a movement along the demand curve to a new equilibrium. Consumers will buy more but only at a lower price. How much the price must fall to induce consumers to purchase the greater supply depends upon the elasticity of demand.
Example 2. A decline in the preference for beef is one of the factors that could shift the demand curve inward or to the left. With no immediate change in supply, the effect on price comes from a movement along the supply curve. An inward shift of demand causes price to fall and the quantity exchanged to fall. The amount of change in price and quantity, from one equilibrium to another, is dependent upon the elasticity of supply.
Two forces contribute to the size of a price change: the amount of the shift and the elasticity of demand or supply. For example, a large shift of the supply curve can have a relatively small effect on price if the corresponding demand curve is elastic. That would show up in Example 1 above if the demand curve is drawn flatter (more elastic). In fact, the elasticity of demand and supply for many agricultural products are relatively small when compared with those of many industrial products. This inelasticity of demand has led to problems of price instability in agriculture when either supply or demand shifts in the short-term.
Thus, price determines how a marketplace operates and price is in turn controlled by elasticity of demand first and then supply.
Besides if you do have any questions give me a call: https://clarity.fm/joy-brotonath

Always focus on the hard part and pull easy part with it, that what I would like to advise you if you have massive resources like money and time it's advisable to try " try and error method " if not you must get data of your business and go with data drove decision. I have created a blog for the same, you can refer it for more details.
https://techwink.net/blog/how-to-solve-chicken-and-egg-problem/

Depends on the Market and Industry. Some industries definitely benefits to release teasers, whitepaper, Ads, & begin stimulating demand in chosen market. But I think the product must be in production before stimulating. I would make sure the product has met every expectation and can be easily replicated in Factory. A good Example is PUNDI-X in Cryptocurrency. They shilled the theory & wallets long before the released their XPOS system to the public. However they did have a few 1000 XPOS ready to go before their Blockchain payment system went live. Slowly they are converting from ERC-20 token setup to their FX chain. Now with FX live & BTC being adopted as a National currency in multiple countries that the XPOS offered by PundiX & FX project will be exploding as more countries adopt BTC as a legit currency. Only company with a Real HARDWARE XPOS that can process any credit card and nearly any major altcoin in existence. They knew that Crypto would be adopted at government level eventually & PundiX postioned themselves to break out between 2021 & 2024. Come 2024 we could see a $50 dollar Pundi-X 7 $50 dollar FX coin. IN the crypto community ICO's(Initial coin Offerings, always stimulate demand a few months before they launch their sale. ) So when it comes to startups, without any merchandise like most ICO's who are just code & Whitepaper with Roadmap. Take XYO for example. They MAXIMIZED their Audience by using Social media and even though they have yet to meet many of their RoadMap bullet points, they were still able to generate considerable demand & massive following long before a single piece of the XY persistent company, or XYO Coin apps went live. So when it comes to creating a marketplace, it always depends on the industry.

You need the supply first, but it should be a sprint effort to get the bear minimum you need to test your PMF then you focus on demand for a longer period of time.