#125 – You can’t make a deal if you’re dead

Rule of Acquisition #125: You can’t make a deal if you’re dead.

To be a donations based organization. You need to be able to accept donations. So we had to put together a campaign to raise funding.

Putting together a crowdfunding campaign isn’t easy. Who are we? Why should people give us money? After a lot of thought and brain storming we finally started realized by looking at what we already had done, that it showed a trajectory that we might be able to guess what we might do in the future from. While we could just be a collection of personal projects, we realized there were some common threads that ran through our projects. We believed in free speech, archival, ease of use, good tools, good design, far reaching goals, and open source software. We used this to form our identity as a group. When people contribute to us, they’re going to be contributing to these ideals.

We created a campaign that put our projects on display through this lens. Everything is tied together through our group’s mission. Then we had to figure out a target goal; what was reasonable?

With my past experience, I realized that I could fundraise about $2,000 myself but Michael had much better success in the past with donations and had a much larger social reach. So we aimed at little under $5,000. If we went over $5,000, that could be rounded up to $10,000 in people’s eyes, if you’re not careful. (see the numerous articles about the Freakonomics/psychology behind .99 cent pricing).

We were already paying over $200/mo for everything, but we knew in about a year that we were likely to double in size at our current growth rate. So a budget of $400/mo. would make this money last roughly a year which was a nice roundish number.

We also had to choose how we were going to fundraise, raise or lose it, keep what you raise in a fixed amount of time, or raise as you go. For what we were doing, the raise as we go made the most sense. We didn’t need a lot of capital upfront to buy hardware, we just needed to pay our hosting bill each money. GoFundMe provided the best set up for this.

We decided to also launch our campaign 2 months before we needed the money, in order to give the campaign time to raise enough money.

We launched the campaign October 22nd and within 30 days had raised 10% of our goal. It was successful enough to see a light at the end of the tunnel. We might be able to continue doing what we love. Our users found enough value in what we were doing to support us. We’re so appreciated of this, I don’t know how to express it.

So now that we had some funding, by launching before 2016, it also lined up with another advantageous event for us. We really wanted to make sure we spent our funds wisely because we’re not sure the campaign would continue to fund at the same initial rate. And we want to be the best stewards of our users’ money as possible and Black Friday was right around the corner.

We used our thanksgiving break to scour all the providers looking for the best deal we could. We created spreadsheets, weighing the pros and cons of cpu, memory and disks config.

We had SSDs with our previous provider. One of the continuing goals of Mixtape is to be the fastest Pomf clone, so we really wanted to continue to have SSDs, however if we sacrificed performance for storage, we could use old slow platters (hard drives) get much more space. Other factors like RAID configuration played into the speed/space mixed. So even if we went with slower disks we might make up for it by using multiple hard drives.

We also debated about location, since we had to consolidated a bunch of VPSes, we’d need one big server and we wanted it to be in the central part of the U.S. so we had good quick routes to Asia and Europe.

So when we saw a dual 2.4ghz octo-core with hyperthreading with 128gb of ram for a steal. So that’s 16 physical cores and 32 logical threads about 2-4 times more than we currently had and about 4 times the memory for the same price. And the disk configuration was just something we could not beat, 9 SSDs drives. Not only were the drives super fast, we’d be able to use RAID to increase the speed further. And with 9 drives, it didn’t matter how small they were, it was still was 10 times more than we had. A find like this quickly draw a close to any discussion or arguments we had over various options.

We ordered it and drew up plans to migrate off our VPSes. Though we did chose to hang on to one of our older servers in the EU because we did have a good deal on it and it had enough space to hold back ups for this new main server.

Through planning and hard work, we continually challenge each other to do better. We listened to our users and community and continue to deliver the best value we can with our resources. This is going to be a requirement if we want to hit those long term overarching goals in the future.


Sapphire Engineering