A quick recap from where I left off in Part I: starting with all of the business ability of a dried plum, I developed an unpopular Twitter tool called TweetingMachine; after nine months, I was ready to scrap in its entirety and mark the whole sorry app as a failed experiment; just before I gave up, a friend pointed me towards ThemeForest, and suggested a couple of themes.
October 2010: What a difference a design makes
I bought the themes, and sat down to integrate them. I was expecting this to take a lot longer than it did: in the end, it took me a few hours over the course of the evening. Bedtime was approaching, and I chose to spend the last hour of the night harassing my ever-patient fiancée with over-enthusiastic demonstrations of TweetingMachine’s new-found greatness.
You see, over the past few months, my hatred for TweetingMachine had built up day by day, its cheery colours and shiny logo only heightening my sense of failure. Thankfully, integrating the themes gave me a new burst of enthusiasm for the project. Suddenly, I was really enjoying visiting the site and playing with the tool. As I woke up the next day, my head filled with all-to-ambitious dreams of wealth and success, and this in turn motivated me to develop yet more features.
Enough talking! What do the figures look like? This is a graph of new free trials:
So, cue wild happiness! But how well did this translate into sales?
Honestly, I couldn’t believe it.
One month, and only five new subscriptions.
November 2010: Running out of excuses
At this point, I was charging $9.99/month, with a free trial of 24 hours. I thought this was a fair deal, but as the stagnating number of new subscriptions showed, something wasn’t working.
What could it be? And why? Of course! The price! That MUST be where I’m going wrong!
Halfway through November, I bit the bullet and made a significant change to the pricing: TweetingMachine now cost $19.99 per year; no more monthly rebills.
I have yet to have any magic moments with changes I’ve made to TweetingMachine. By that I mean changes that have had an instant effect on use or payment. The pricing change was no exception… for the first 2 days.
It was now the middle of November. And on the 15th, someone subscribed. On the 16th, another user paid. Teasingly, no-one signed up on the 17th… but on the 18th I received two new subscriptions. The pattern continued for the rest of the month, averaging one sale per day.
Well! This was MUCH more like it! November brought in over $200!
December 2010: What was that about marketing?
The rate of new subscriptions continued throughout December, sometimes two subscriptions a day, and on one memorable day, five users subscribed!
So now the concept’s proved, how can I get more potential users to visit the site? I took the view that if in doubt – as I was and continue to be; note earlier implied reference to business ability of a squashed frog – follow what your competitors are doing. So, I sat down, typed my competitors’ names into Google, went through page after page of links, and identified bloggers who might be interested in covering TweetingMachine. I sent hundreds of personalised emails, and received under ten reviews in total. Oh well, better than a kick in the teeth.
I also found lots of directories of Twitter tools: type in a description; upload screenshots; get listed; lots of happy users dance their merry way towards your site.
I was expecting new subscriptions to tail off just before Christmas; it made sense to me that there might not be that many people online, and even fewer ready to hand over their hard-earned cash after the yuletide spending craziness. Imagine my happiness when new subscriptions continued, including on Christmas day itself!
December brought in just under $500 – to be precise, $479.71 after PayPal fees.
January 2011: Brave new subscriptions
The first month of 2011 was a time of high emotion. For several days on end, no users would sign up… and then a flurry of three or four subscriptions would come in within a couple of hours. My mind even deluded itself into thinking it was acceptable to describe this as “A rollercoaster of emotions.”
Whilst I was happy with January’s takings – $429.75 after fees – the new signups permanently tailed off at the end of the month. Something was definitely amiss.
February 2011: Delusions
I had started a new job in January, developing Facebook applications for a local startup, and this was taking up an awful lot of my evening time. I had some ideas for changes to make to TweetingMachine that I was keen to implement, but wasn’t sure when I’d get around to it. After all, assuming that my time is free, each month it still bought in the equivalent of a few nice meals out.
Still, subscriptions had nearly stopped coming in altogether; for reasons unknown, TweetingMachine had six people in total subscribe in February. I had strangely depressing thought: if I had been lucky in December and January, quite how much money were those who knew what they were doing making?
I eventually found the spare time, and made a couple of small, but effective changes: I increased coded some flexibility into prices, so future price changes would take seconds to implement; and I did the same for the free trial period.
Going forwards, TweetingMachine cost $19.99/month, and the free trial increased from 24 hours to 10 days.
March 2011: All change!
What difference did increasing the price 12-fold make? Colour me shocked, surprised and, frankly, happy: it made *zero* difference!
Actually, I tell a lie; the rate was between that of January and February. Essentially the same… except that these subscriptions were going to be rebilled each and every month!
I was starting to feel cautiously optimistic. There were still plenty of outstanding questions (such as: how do I get MORE MORE MORE users to visit?), but for now my most pressing question was: will these subscribers continue their subscriptions next month?
April 2011: The answer is… yes!
Please forgive me if this article feels like a con; I’ve had months where I’ve made just under $500, and months where I’ve made a lot less than that. April’s on target to make over $500, and as of the 27th, I’ve had one unsubscribe from the previous month, out of 17 new subscribers. Not the greatest retention rate, but if that continues for the next twenty years…
Going forwards, there are a few tasks I need to get on top of. Number one on my list is to get into A/B testing. I’d always dismissed this due to: a) laziness and b) incompetence. Well OK, since I’m being honest here, also c) I thought that I didn’t get enough traffic to be able to draw any meaningful conclusions. I believe that TweetingMachine now receives enough traffic to warrant proper testing, and I’ll be sure to write an update post about how I get on.
Now, this isn’t a story of huge, wild success; it’s of a 29-year-old making his first steps in business. I believe that in a few more years, after a few more failures, and a few more, modest, successes, I’ll be in a pretty good place for my first major success. The point is: you have to make a start.
What I have written here has been a lightning-quick history of TweetingMachine; if you have any questions please write a comment and I’ll reply as soon as I can
I have a few further blog posts lined up, including the following:
- The technology I use for my projects, and why.
- How I integrate themes from ThemeForest into my apps.
- My other projects, including FasterDev and InboxCleaner.
- How I made a reasonable living freelancing on sites such as vWorker (previously known as RentACoder).
If you’d like to know when any of the above are posted, you can subscribe to my updates list using the form on the top-right of the page. If you wouldn’t like to know – tell me why, and ideally tell me what you *would* like you read instead