Web Development Blog

Some Background:

OK, so I thought I’ll just give you some background 1st so that you know what I was trying to achieve by this whole process. Also, let me just make it clear that I will be vague about certain details as I don’t want this to replicated down to the username.. that might cause some spam bells to go off at Twitter, but I’m sure that if you know Twitter well enough you will be able to replicate the results easily.

The service I am promoting is a site I built about 3 months ago, and the basic function of the site is to publish a fact and a blurb about the fact and a link to the full article.

I haven’t really marketed this site at all, so the primary source of traffic at this point is from people who are running the site’s widget on their site, which isn’t as much traffic as you would expect.

So here is what I have done in the last 3 days and the type of results I have seen as a result:

Day 1:

On the 1st day I created the account on Twitter that I was going to use for the experiment. I decided to go with a relatively descriptive name (as far as possible of course).

After I created the account, I decorated the page slightly which basically consisted of a custom background and matching color scheme. The reason for this was simply to make it look more legitimate to new users that will check it out. First impressions are lasting!

I then added 3 Tweets to the page in the same way the rest of the tweets will looks, so that it doesn’t look like one of those empty spam pages.

On the site part of things I then created a RSS feed specifically for this. What the feed consists of was basically a singe item that updates randomly on every load with the current date and time. This might seem slightly useless, but the next step will explain this…

Once all the above was done, I created an account on Twitterfeed and loaded the above feed into the service. I set it up to post the latest feed item every 24 hours with the title of the RSS item as the primary tweet and then the accompanying link to be posted as a TinyURL link.

I went with TinyURL simply because it’s widely used across Twitter and therefore is trusted by users. The second reason I went with it was so that the users can’t immediately see that all the links was pointing to the same site.

There we go, primary setup was done and ready for the big quest for followers on day 2!

Day 2:

The main purpose of day 2 was to follow 500 people and see how that translates to people who follows you back, simply because you follow them.

Finding people to follow was interesting, I decided to go with the big names on Twitter, simply because there are large numbers of people to follow without any effort besides click, click and click.

I chose big names of people who would find my type of service interesting, for instance, if you are going to be posting scientific data daily, then Oprah or Britney’s followers are probably not the best place to go.

So, I found my main guys and started following their followers.. I tried to not do more than 100 follows per hour as I didn’t want and spam bells to go off on Twitter’s side.

About 5 hours later I had 500 people on my list of people I’m following and within about an hour or two I had about 40 followers.

A day later I was on my grand total of 95 followers. It stopped at 95 with the off one adding me here and there and the odd one blocking me.. but it’s pretty consistently on 95 followers now.

That is a return follow percentage of 19%, not to bad if you would start working with really large numbers!

Day 3:

OK, so day 3 was the big day for seeing how this translates to traffic when we send out the 1st tweet containing a fact.

I added the Statcounter tracking code on my site so that I can see in real time where visitors are coming from and how many of them there are. You can obviously use your own tracking service, I chose this one because I’ve been using it for age and I feel I can trust the information it provides.

When the 1st Tweet went out from Twitterfeed I was waiting to see if there will be any significant rise in site traffic, but what I saw was what I expected, there was very little, to be honest I only received a single visit from this source.

If you would round off the numbers, only 1% of my followers clicked through to visit the site.

Final thoughts:

The results was hardly groundbreaking, but I think it gives us valuable information as to the amount of numbers you will need to make a significant impact on your site’s traffic using Twitter.

I’m also sure that this can be tweaked to have more impact and I would love to hear your thoughts on this, so please leave your replies and ideas in the comments below.