Curation vs. Creation. How to be Content with Your Content
UPDATED, Aug 10, 2016: Recently, I did a deep-dive into Twitter's API. Ok, so it was more than just a deep dive. I ended up writing a complete Twitter engagement and growth-hacking system.
That's not what this post is about. I thought it would be a good opportunity for a social experiment.
I tested the platform on three accounts. Mine, a client's, and a friend's company's account.
My account is quite a bit larger than either of the other two. Let's forget about it for now.
My friend's company had no Twitter presence. Here, I'll call it FriendCo.
My client had 409 followers. I'll call it ClientCo.
I thought it would be a good opportunity for a social experiment.
You see, FriendCo does not create content yet. They just retweet tweets from a list that my system identifies as popular.
ClientCo provides a regular stream of high-quality, targeted posts. They do that weekly. I know their posts are of high quality because I write them (LOL, I'm shameless.)
This is a classic Curation versus Creation scenario.
As a content creator, how could I pass up the chance to study it?
Here's what I did.
First, I wanted them to start from an equal footing. I held off promoting ClientCo until FriendCo crossed 400 followers.
Both FriendCo and ClientCo used the same system at the same times.
The main difference was that one curated content the other created content.
My KPIs were follower growth, number of mentions, and number of retweets.
People often add profiles to Twitter Lists. They do that because they want to follow them more closely. It's a high form of engagement.
So, I also looked at the number of lists they were on.
I gave all KPIs the same weight.
So what happened?
Well, at first, curation pulled ahead of creation. It pulled ahead fast.
To say that that surprised me is an understatement. I dug into the "why" of it.
I should mention that I based my Twitter system on the "influencer marketing" concept. It turns out that the influencers we selected for ClientCo weren't all that influential.
Worse, they didn't match the target readership we had planned for.
We changed them, but not before FriendCo pulled far ahead.
After two weeks, FriendCo's curation strategy was ahead 734 followers to ClientCo's 597. More important, retweets and mentions were nearly 2:1 in FriendCo's favour.
I was starting to question the usefulness of my work. I also wondered if I would be able to draw any meaningful conclusions from the experiment.
I worried for naught.
As I write this, a month into the experiment, FriendCo is up to 1018 followers. It earned 174 mentions and 58 retweets.
That's not bad for just clicking buttons.
ClientCo, despite the slow start, is at 1645 followers. It earned 281 mentions and 179 retweets.
FriendCo is on 13 lists.
ClientCo is on 68 lists!
I often wonder how much different it would be if we had started ClientCo with the right influencers list. Oh well, setting the right influencers is part of the process.
So are stumbles.
This not, by any stretch of the imagination, a scientific study. It isn't meant to be. Still I think it's safe to draw a certain conclusion.
It's better to distribute your own quality content than someone else's.
It also avoids the embarrassment of promoting your competitors by accident.
To be honest, I did not expect the difference to be so pronounced. All that said, curating content has the advantage of being much cheaper.
This experiment also outlines an important point. ClientCo's initial stumbles were not due to the content. It was due to the promotion of that content.
Creating the content is only half the story. You still need to promote it to the right readers.
Update: August 10, 2016
Some people have asked how things are going now, 6 months after this post was originally published.
Not very well, I'm afraid.
I probably should have chosen different subjects for this "study."
A little while after this was published, FriendCo hired a full-time social media manager. She asked to have the Beast system connection cut from their account.
FriendCo went from 0 followers to 1018 in a month and a half. This, with only selective automated retweets for content.
Six months later they are at 1209.
She doesn't post new content and she doesn't curate. She doesn't even use images in her tweets. Don't ask me what she's doing. I have no clue.
Not my problem.
ClientCo found I tweeted too often. I would tweet every new post 10 times a day for the first two days, and 6 a day for the next week. Then I would drop to 3 a day for two weeks. Finally, I would tweet twice a day for a month.
That's not much at all.
ClientCo asked that promotional tweets drop to no more than 2 a day for the week a post is published.
People follow people they see. A single tweet is an exercise in futility. Two is not much better.
Another big mistake, but theirs to make. I write the posts. I promote them as a value-add. No worries if they don't want the promotion.
More importantly, they decided they didn't want new content over the summer months. A very big mistake. Slowing down is OK. Stopping is not.
Engagement dipped tremendously. So did follower growth. It actually started backsliding.
A growth rate of 700-1000 new followers a month and mentions up the wazoo went pfffft.
My account is chugging right along. I started on Twitter in June of 2015. Six months ago I was at 4800 followers. Now I'm at 14,719. Better still, I've flushed most of the untargeted ones.
I'm left with a very busy, very supportive following.
I'm on 774 lists and see about 500,000 - 700,000 impressions a month. I have no complaints.
I also wrote a lite version of the Beast. That one is as yet unnamed and is in beta-testing. It carries most functions that the Beast does and adds a few as well. It's light on growth hacking, though.
That will come later.
If anyone would like to join the beta-testers, just say so in the comments. The more the merrier. I expect it to go to market within a month or two. That is if I ever figure out what to name it!