TylerCruz.com: An Internet Entrepreneur's JourneyThe journal of a web entrepreneur on his journey to wealth.

My Case Study on TwitterCounter Follower Advertising

A few years ago, I posted a couple of case studies when I used the service Twiends in an effort to boost the number of followers I had on Twitter.

Fast forward to today, and I’ve managed to keep a very high retention rate from those followers. In fact, I currently have 92,500 followers. Unfortunately, most of those followers are not very active or engaging – my tweets do not receive that many favourites, replies, or retweets. The followers from Twiends were indeed real Twitter users (not bots or fake accounts), but the engagement was just low.

And so, a couple of weeks ago I set off to look for another source to gain more Twitter followers, this time wanting the focus to be on quality over quantity.

The first place that came to mind was TwitterCounter – I use the site fairly regularly to view a graph of my follower growth over time, and was often intrigued by the "Get Followers" section they have on the site. And so, I decided to give it a try.

Purchasing The Traffic

I opted for the one-time package of 25,000 impressions, of which an estimated gain of between 230 and 280 new followers was given, for a price of $150. That would work out to around $0.60 per follower which is a hell lot more than Twiends costs, so I was really hoping for quality followers.

My goal was for 1% of those new followers to be active and engaged, meaning that 1% of those new followers would either reply, retweet, or favourite each of my tweets on average. That works out to between 2-3 engagements per Tweet. That was the goal, as that is easily worth $150 to me.

In order to properly run this case study, I decided to run the campaign for the Twitter accounts on one of my other websites, Movie-Vault.com, as its Twitter account (@movievaultnews) only had 76 followers with no existing engagement, so I would be able to accurately measure any future engagement properly (since my blog’s Twitter account does get some engagement already).

When ordering the campaign from TwitterCounter, there were a number of settings that I could choose from, such as spreading the traffic over time or having it go all at once. I chose the latter option.

One of the options was a bit sleazy; for the cost of 2 credits (impressions), you could target a specific country, otherwise you would receive worldwide traffic. So right out of the gate, if you want to avoid gaining followers who may not even speak English, you have to pay double the amount you initially thought. I decided to go with the global traffic.

The Wait

For some reason, I was under the impression the entire time that upon paying and setting up the campaign that I would receive the traffic quickly; I thought it might take at most a day to approve the campaign, if it didn’t start instantly.

Part of this was due to the fact that the site stated "Estimated Start Date" to be November 11th, 2015 (the date I ordered the campaign). I realized that it was a holiday and so checked back the next day. The traffic still hadn’t started and the "Estimated Start Date" changed to November 12th, 2015.

A few more days of this and I decide to reach out and contact their support to ask when it will start. I get a cold response saying that it could take a week and that the system will e-mail me when it starts.

Fast forward a week and traffic still hadn’t started. On the 8th day, it finally did start, although I never did receive an e-mail notification that it did.

In the screenshot below, you can see what the campaign dashboard looks like while it’s running. In addition, you can also see where and how your ad appears on their site:

Waiting 8 days for the campaign to start wasn’t so bad – it’s the fact that I had no idea when it actually would.

But it gets worse: it ended up taking 6 days to deliver the traffic. I had ordered the smallest one-time package they had. If I had ordered one of their larger options, such as around 2,000 followers for $800, then it would have taken 80 days to receive the order (assuming the same rate held up).

This makes the whole "one-time" package a bit misleading, as I had ordered their smallest package on November 11th, and received my order filled on November 25th.

Okay, so it’s not instant – not a huge deal. My goal here was quantity followers, so let’s focus on that.

The Results

Here’s how the campaign looked like when completed:

The conversion rate of just over 1% held throughout the campaign and TwitterCounter ended up delivering me 263 new followers which was right smack dab in what they had predicted and advertised, so they get a thumbs up for that.

So what about the initial question of how good the quality of their traffic is?

Unfortunately, from the small case study I did, it’s pretty crappy.

The @movievaultnews Twitter account automatically sends out Tweets whenever a new news post is published to the site, which has recently been around twice a day.

The screenshot below shows the typical response in engagement the average Tweet would receive:

To clarify, that’s around 28 "impressions" which is meaningless – there are no expands, clicks, retweets, favourites, or replies.

And some of the Tweets were a bit interesting too, so I don’t think "boring Tweets" were really a factor. For example, one of the Tweets was "Toy Story 4 Leak Confirmed" with a link. That Tweet did end up receiving 2 retweets, but upon closer inspection I realized that those were because I retweeted it myself on my blog’s Twitter account and the other retweet was from one of my blog’s Twitter account’s followers – not from @movievaultnews!

Also, I am not sure how Twitter counts engagement… for example, if I click the link twice does it count as 2 link clicks even though it’s from the same computer?

Because of this, I am not sure how much of the stats I accidentally manipulated by engaging with myself:

Below is a graph from Google Analytics showing all referral traffic from Twitter. Remember, traffic didn’t start until November 19th and some of those clicks were from me:

In the end though, there seems to be no doubt that despite gaining 263 new followers, there was no noticeable gain in engagement from followers.

I guess my case study would have been better if I had ordered a larger campaign for a larger sample size, but if these results are any indicator, then it would have probably just been more money wasted.

I may try running a campaign using Twitter’s very own built-in ad platform and see how that goes, and publish a case study on that.