The details (tactics) are likely different based on who your target customer/user is - i.e. different for b2c than for b2b, but the overall process is the same. Get 10, then get 100 then get 1,000 then get 10,000. By the time you have 10,000 your process (tactics and strategy) should be pretty well defined and repeatable. the firs 10 - these are usually friends, family and colleagues. They are users because the know you and like you. Get the next 100 - these are often acquaintances of the first 10. Study in detail how that process worked by talking to everyone of these first 110 users/customers and really understand why they are users, what value proposition resonated with them and how they discovered your product? Don't skip this step - talk to them, preferably face to face. getting from 110 to 1,000 is more challenging. Do you simply replicate the processes that worked for the first 110? IF you have something that is viral then you've struck gold, but don't bet your company on only that same process as the discovery process may be different, but the value proposition is still likely valid. this is where instrumenting the process pays off - you don't have time to have a meeting or phone conversation with each of these 890 new users, but you still need to know how they discovered your product and what value proposition resonated with them. Again, the tactics are different (most likely) if you are direct selling to large businesses VS marketing to consumers, but companies like Slack are proving that the tactics don't have to be that different.That process - those strategies and tactics that got you from 110 to 1,000 is likely a good process to repeat to get you to 10,000 and your instrumentation should give you good data to tweak the process if needed. The process could be some combination of Adwords, FB ads, social, content marketing, events, influencers, direct sales, email marketing... The important part is to 'instrument' the processes so that you have data about what works and what doesn't for each $1 of input. If you could have a conversation with each of those first 1,000 or 10,000 users/customers you would quickly see the patterns that work so how do you 'have a conversation' without actually picking up the phone? It could be A/B testing, it could be surveys, it could be user ratings (i.e. Amazon/Yelp), it could be email, it could be google analytics, kissmetrics, Mixpanel, or some combination. The important takeaway is to have conversations early on so you make some educated assumptions about processes and value propositions and a basic understanding of what to instrument and how so that when you test new tactics you can actually see what works and what doesn't VS guessing. All of this, of course, assumes that you can deliver on the value proposition that users are attracted to and support you existing customers as you grow - that's an entirely different conversation, but equally important.