How to price: Lessons learned from our first attempt
blog.filepicker.io
blog.filepicker.io
A few months ago, I was launching an information product I had been working on for a few years. I was trying to decide how to price the product. I was planning to price it anywhere from $17 to $97.
In order to come up with pricing, I did some testing before the launch. I put up a sales page that wasn't linked from my main site. The 'Buy Now' didn't work on the page (It sent them to an e-mail sign up page instead).
Then I sent a ton of targeted adwords traffic to the sales page. I used Google Website Optimizer to split-test the page with a range of different prices: 17, 27, 37, 47, 67, and 97. Google Website Optimizer allowed me to see how many people clicked 'Buy Now' when they got to one of the pages. What I found surprised me.
The highest number were at 17 and 97, and were almost equal in number. The other prices 'sold' less. This to me meant that people were willing to pay 97. (I could have possibly gone higher but didn't want to charge my readers more than that.)
So I ended up selling the product for $97. It's been selling pretty well. After the launch, when the product was actually for sale, I did find out that people who click 'Buy Now', don't always buy. In fact sending targeted traffic to the page hardly sold any at all. A certain number would click 'Buy Now', but no one ever bought. Only people who came from my website bought the book.
Does this mean that my tests were off? No, I still think it was the best way I could have tested. A few weeks after the launch, without announcing it, I split-tested the real sales page for two weeks with my readers by selling it for 47 and 97. So at random, buyers would see 47 or 97 and could actually buy the product. The product sold almost an equal number at each price.
Not to mention, the added benefit of the pricing plan as it stands is that I'm literally thinking in my head "I hope I have to pay for this someday soon!"
However, as we went on, we kept needing server side components to make things work smoothly. For instance, we handle large files upload by chunking them up and reassembling them, retrying parts as needed. Image conversion baked in and other components in the pipeline that we want to build also require servers, something that you can't do with a plugin.
We fully support plugin developments and we fix bugs or add features as we see them.
When we first launched, we had people emailing us asking to pay and we've been growing week on week so the developer need is there. Part of me is just as surprised as you are, until I write some code dealing when multipart form uploads or box's inconsistent error codes. Then I remember.
Yeah I was pretty skeptical when these guys first launched. I mean why not just DIY?
But if you think about it doing file uploads across tons of services and platforms is a PITA. As much as we love to tinker and hack, when you do need to execute fast, there can be a downside to reinventing the wheel.
@filepicker - Congrats on the the new pricing model, it's a definite improvement. Free is always a great price!
Also given that files are stored on customers buckets I feel is a good move, as it prevents lock-in.
It does offer a lot more than any jQuery plugin does though, and doesn't require any server-side code. File upload is a huge PITA to begin with, not to mention integration with Facebook, Dropbox, Instagram, Flickr, webcams, etc, etc.
You're sending messages without an opt out which could get your domain blacklisted in the long run. I'm not saying I minded, but just be wary :)
Great article btw.
Where are the files stored? We automatically save all uploaded files directly into your Amazon S3 bucket. To make it easier to get started, if you haven't put in your S3 credentials we will store them on our servers, but as your usage increases we will ask you to move to your own storage.
People have been asking for extensions as well, so that's on the roadmap as well.
We have some hostname verification and we also also adding in secret keys to sign requests so we can be even more sure.
We also have some checks that look for abnormal upload patterns that have found a couple oddities and will get better with time.
Looks more like $25 then, what's the catch? :)