I hate seeing startups fail, even though I know that it's part of the game and a reality that all of us endeavour to bypass every day. I also appreciate this post-mortems (even though most of them do it wrong), since it's always a great opportunity to learn (from others' mistakes and / or experiences).
As I read these articles though, there's a couple of take-aways (mine) that wasn't overly obvious and I wanted to share them here:
Freemium is great for traction, but hard for building a business. You are up against the odds, when trying to convince a free user to pay you. Some might say this is mostly impossible, if not at least very unlikely.
Growing a business on a +- $5 per month revenue will always be a challenge. In terms of expenditure, how much can you buy for $5? This also begs the question to be asked: Would paying customers have paid even more to keep the service alive?
When you are working with longer product roadmaps (6+ months), you need to be aware of the fact that you may reach the end of that road and yet there won't be guaranteed revenue or profitability. Regardless of how much you build->measure->learn, the bigger the thing you have to build to generate sufficient revenues, the bigger the risks involved.
Customers (and revenue) will always be your best investors and you don't have to cede any kind of control to them in return. I get that some businesses simply requires initial capital / outside funding, but nothing in life is free; especially not a $2m funding cheque.
I'm not an expert in investments and valuations, but I do understand the basics. In Everpix' case, they raised 10x more in funding than they ever made in revenue. So we're talking of a P/E ration of more than 10:1 (depending on how much equity they forked out for that funding), which is incredibly high.
Looking at what would be deemed as a success for Everpix, could they have shifted the business to focus on a smaller audience, who were willing to pay much more than the $49 per year for the service? Could they have built the same product for 1000 or 2000 people and had been profitable? (I don't know if this would've been viable; I'm merely asking this as a hypothetical question.)
There's more than one way of skinning a cat and none of my take-aways above are to point the finger at Everpix. If anything, I know startups are hard and there's no perfect practice to make sure every startup succeeds.
But because there's more than one way of skinning the cat, it's always good to consider the alternative(s).