My Twitter Automation Experiment. Part 4: 90th Day

By | November 29, 2015

This is the fourth article from the series about my Twitter automation experiment. Like I mentioned in my previous blog posts, I am starting an experiment to see if it is possible to grow a Twitter account by spending no more than 30 minutes a day and get a good quality of traffic to my blog. Every 30 days I will be writing reports about the results I achieved and conclusions made about this experiment.
This is the fourth part of the series. If you haven’t read the previous part, it can be found here:

This month was a bit better than the previous one. Everything is growing steadily without any significant jumps. A little faster than last month, but not as much as I was expecting. I already have some ideas how this growth might be boosted, but I will talk about it next month after I’ll test everything and have first results. Also, I already started split-testing auto DMs. It is still too early to jump into any conclusions, but some assumptions could already be made. I will talk about it later in this post.
Unfortunately, I had some problems with my hosting service for a few days this period which had some impact to my performance results. Luckily, we managed to solve the problem and it should not repeat again.


What I managed to achieve?


As you can see the follower list is growing quite steadily every month. I already reached 3028 followers by this day. It is 1030 people during the 30 day period and 50 followers more than last month. So far, this is the best month of all three and I feel very positive that we will see even better results next month.


All the main interests of my audience are keeping top position for the third month in a row. It means that I am targeting the right people who should be interested in my content.


I am glad that the percentage of my audience from US and UK are keeping steady. They consist more than half of my audience. Hoping that together with Canada and Australia it will reach at least 70% of my total audience very soon. Also, it would be great if some top European countries could join this audience in the future as well.




As I mentioned at the beginning, I had some problems with my hosting service which stopped all the Twitter automation process for a few days and it had a little impact to the results. But results are still good.

Despite the steady growth of followers, the average number of daily impressions wasn’t growing so fast as I expected. And, because of those few lost days, I managed to reach 2100 impressions a day. It is 200 less than last month.
However, I am glad that the number of link clicks increased by more than 40% and my tweets received 126 link clicks in total. The CTR increased from 0,13% to 0,19%.


Traffic and conversions


Luckily, the traffic has increased by 14% compared to the last month. I managed to reach 113 visits during the last 30 day period. The number is still lower than the first month (130 visits).

Because of the Twitter auto DM testing, my conversion rate has decreased. This happened because fewer people received the message with a link to my free e-book. However, the conversion rate of this direct message stayed the same – 18%.


Auto DM split testing

During this period I started testing different Twitter direct messages to different people. I created 3 totally different messages and sent them to my new followers:

  • The first one is an introduction to myself and invitation to download my free e-book
  • The second is just a simple greeting with a link to my home page at the end
  • And the last one is an invitation to read my article about free traffic generation methods

I am still leaving this process for another 30 day period without any changes, but we can already make some early assumptions. As we can see the performance of the second message is very weak so far. I received only 4 visits to my blog. However, the first and second messages are showing some interesting results. The first one is still performing best and has 10 conversions. But the third one has a much better average session time on my blog which means that they are reading my article.
It is still too early to make some adjustments to this test, so I will need some more time before making any changes. Probably I will do it after the next period ends.



I already have some ideas how I might be able to boost the performance of this Twitter account. I am hoping that I will manage to implement everything very soon.
As previous 2 month showed, the growth of my Twitter page is not as fast as I could have expected, but I am satisfied with the results I managed to reach this month. It is slowly getting better and better. And more people are being directed to my blog. We’ll see what next month will bring because it might be hard to reach any better results as Christmas and New Year should impact overall results.

But, best of luck to us all!

And, if you have any questions or suggestions fell free to contact me. It is always a pleasure to hear different opinions.


Read more:

My Twitter Automation Experiment. Part 5: 120th Day