SaaS #2 | Thoughts on Building Marketplaces
Supply v/s Demand. Or how do you chicken the egg. Or egg the chicken.
In my last post, I argued that marketplaces are flywheels of value creation.
In this post, let’s look at the process of building one. Or to be more precise – which side of the marketplace to focus on. We are going to take a more generic PoV here.
Short answer – both, but at different times.
Supply is critical when you start out. Long term, your only strategy and enduring moat will be how well you can aggregate demand.
1. What’s my constraint?
Here is a key thing that gets lost in translation - ALL SUCCESSFUL MARKET PLACES ARE CONSTRAINED, at least initially.
Why? It’s a better strategy to be a bigger fish in a smaller pond. Instead of stretching yourselves thin, focus on one area, ace it and then move on. I’ll let Frank Underwood explain it best:
But why is this important? Because supply is closely tied to Geography and/or Category. If you try to be everything to everyone at once, you will likely fail1.
2. Building Supply
To get a customer, you need to build supply. Now, there are multiple ways to go about it. Here is a small graphic to outline a few ways to go about it. You will see that the successful companies have a mix of these strategies. As an example, OpenTable started off with FoS selling their standalone tool and referrals. Depending on what you are trying to build, your strategy mix for supply building will evolve. Here is a graphic. Please excuse the potato design, not very comfortable with online tools yet. Also there are a couple of typos but I cant be bothered to fix them.
However, your suppliers won’t stick around for very long if they are unable to get business of the platform (subsidies notwithstanding). So now that you have a critical mass built up, all attention needs to focus on to building demand.
3. Building Demand
I would posit that aggregating supply is not the hard part. Aggregating demand is. As a supplier, if its not costing me any money and there is potential for money to be made - of course I will sign up. Whether I would stay on or not is a different story.
Marketplaces are powerful. But each time a new marketplace is built up, it is not really reinventing the wheel or discovering a new universal constant. It is about doing better on one of the 3 value creation axes - Cheaper, Faster or Better.
Once again we go back to Metcalfe’s law. Now, here is the thing - there is always an incumbent way of doing things. Unless your customer (demand) is not happy with the experience - they are not going to come back. In the early days, the inertia (which will eventually help you build your value creation flywheel) is working against you. The inertia of moving away from status quo is surprisingly high.
A great example is Uber. It took business away from taxis. GOAT beat ebay and craigslist on Sneakers.
Why? Because the friction was lower, perceived value was higher, experience was better. Customers were happier transacting on these platforms.
Now let’s go back to my point about constraining - See what I am getting at? Smaller areas allow you to optimize your offering and give a better experience. You will find the next opportunity and sequence from there. NPS scores are a good proxy. You need to come to a minimum baseline level before expanding into adjoining areas.
Friction is the happiness killer2. On both sides. On the supply side you may be worried about how your CAC compares with LTV (or is the juice worth the squeeze). On the Demand side, friction may come from trust issues, good matching, level of poorly converting searches and so on. This will take time and is a never ending process. My limited insight here is that this is not something that can wait. Reducing friction is an extremely important part of successful demand aggregation.
Now, you may be wondering why there isn’t a very good infographic like the one above yet. It’s coming. I wanted to take the time to touch on happiness because it is the single most effective, sticky way to build demand. Word of mouth is extremely powerful. Slow, non-scalable even, but powerful and sticky.
Now for the infographic:
One final thought - Loops are not covered here. I don’t know enough to talk about them intelligently. Mostly, I somehow don’t see loops here. Maybe I have been out of it for too long.
Key is also to figure out whats not working and move quickly away from it. Think of a failure standard. If X strategy has not delivered Y metric by Z date / spend, we will move on to something else.
As before, thoughts and comments are welcome. Next post on marketplaces will lay down some thoughts on scaling.
Link to second post in this series on Scaling.
There are notable exceptions to this. ThumbTack did everything everywhere, effectively scaling and getting off the ground simultaneously. But only example I can find of this is Thumbtack. Standard disclaimers about single sample sets apply.
Reference - Dune by Frank Herbert. Paul meeting Bene Gesserit for the first time.