Find your voice and get your story straight
Using natural selection for a Twitter cull.

Using natural selection for a Twitter cull.


Culling is usually only required when natural selection, or more precisely nature itself, fails to control relative populations of species in an ecosystem.

It’s either natural selection or culling.

One or t’other.

Well I’ve used both to reduce the number of people I follow on Twitter.

One of my favourite ever clients, a lovely lady from Belfast, used to tell me to “catch yourself on” whenever I wasn’t getting it in her eyes.

And I had a “catch yourself on” moment earlier this year when thinking about Twitter in the context of me in social spaces.

It’s so so easy to get sucked in to watching and nurturing your follower count. But when I stepped back and thought about how Twitter has worked, when it has worked at its best for me, it has very little to do with quantity and almost everything to do with quality.

Following the right people (regardless of whether they follow back).

Sharing the right kind of content from the right kind of people.

At the right frequency.

Having the right conversations.

Building relationships beyond Twitter, be that at real life events or via RSS and blog comment threads.

Indeed I wrote a post back in January about the relevance of RSS to me, which got me thinking about the strength of ties in social spaces rather than just the number of ties.

Cutting to the chase, I’ve reduced the number of people I follow from 968  to 770 in the space of a few weeks.

Here’s how that process looks when charted by Twitter Counter.

That’s a 20% reduction in the number of “friends”.

A 20% cull.

In fact the actual number culled is higher than this because I’ve continued to follow new, interesting people throughout this period. 20% is the net reduction.

And, as mentioned above, I did this the hard way but the right way.

Natural selection, for which read quality control or survival of the fittest for purpose.

Natural selection was a relatively painstaking manual process, as a result of which I’ve had a close look at every single person or organisation that I was following.

Would I miss their content?

Have I “engaged” (ugh!) with them?

Was I likely to want to engage with them?

In some, but very few, cases were they still active?

I never made a decision to unfollow based on whether anyone was following me.

It took quite a while but it was an enlightening and highly worthwhile exercise.

No disrespect but God I was following some dross.

The proportion of practitioners and originators amongst the people I’m now following is much higher than it was.

The proportion of theory-mongers and self-styled gurus much lower (tending to zero in the case of the latter).

I assumed that I’d suffer a significant reduction in the number of people following me as a result of this exercise, on the basis that a fair few people would be actively tracking and monitoring and maybe even automatically reciprocating for unfollows.

(I’ve never been bothered with that level of “sophistication” myself.)

Here is what has actually happened.

The scale of this graph is not the same as the one above but there has been a 1% drop in followers from the peak to today.

20% reduction in “friends’, 1% reduction in followers. Anyone had a similar or markedly different experience to this?

This has been a valuable exercise on a number of counts.

  • It’s no bad thing to step back and think about what you’re doing in these spaces once in a while.
  • The laborious process of manually reviewing everyone I was following was enlightening and instructive in all sorts of ways.
  • This naturally selective cull has acted as a catalyst for a number of other developments. These include some thinking on how I can make better use of Twitter lists (I’ve tended to use Tweetdeck columns as a surrogate for lists proper). And I’ve had a similar spring clean of the blogs to which I subscribe via RSS.

Everyone has their own personal answer to “what is Twitter for?”

And that answer probably evolves over time.

That evolution can lead to a discrepancy between your objectives and your actions. It did for me anyway.

Have you asked yourself the question recently?


  1. @bureauista

    I’ve had a much bigger cull recently. I confess I do periodically check to see who is following me back and I will sometimes use that to help me decide who to unfollow, but there has to be another reason as well.
    I’m increasingly unsure what Twitter is for. I use it more as news source than a social tool these days and I miss the old days when there was a ready source of banter. I guess I’m still uncomfortable with the divides between professional, personal and private, so I stay quiet and ponder more.

    1. Thanks for the comment Kirsten. Getting that personal/professional balance right is increasingly tricky. The old Twitter = professional, Facebook = personal, black and white view of the social world doesn’t cut it any more. Things like coffee mornings, tweet-ups and Twestivals mean that professional, virtual acquaintances evolve into more personal relationships over time. For me though Twitter remains pretty much exclusively the domain of people that I’ve met/followed through or because of work. I can understand your pensiveness. There was a time when, as a follower of both of you, I “overheard” some of the Twitter banter between you and @infobunny. Whatever happened to them?

  2. ScottB

    I’ve noticed quite a large amount of bot accounts automagically following me for a few days and then unfollowing (presumably if it’s unreciprocated). One digital agency followed me with 10 staff accounts for a week before all unfollowing which I found a little bizarre. Something like 40% of the followers I’ve had in the past month have been bots or random followers who quickly unfollow if you don’t reciprocate.

    And sadly, my local branch of Domino’s started following me after I tweeted about them sending out 11am texts on the day of the Royal wedding. Is that the digital equivalent of inviting the delivery guy in for beers when he brings the pizza? I feel like Adam Sandler in Big Daddy, but less comical!


    PS Phil, I emailed Blonde about the DP job going but haven’t heard anything back. Has the position been filled?

    1. Thanks Scott. Avidly following follower numbers is a slippery slope to desperate measures I think. Like you say, so much of it is automated and “inhuman”. I have a close look at everyone who follows me and apply the same criteria as I did during the cull to decide whether to follow back. It’s easy to spot the bots and timewasters. Introduced by @allanbarr, I had a brief Twitter conversation last night with one @treypennington. It’s clear that my laborious, manual evaluation and culling approach wouldn’t scale for someone like him who is following over 100,000 people.

      Sorry about the Blonde email. If that was via the website, they go to Pete our MD who is usually very good at replying or forwarding to the relevant people. I’ll chase that for you today.

  3. I must say Phil this really is food for thought.

    Whilst I’m an ardent fan of lists the number of people I’m following as a whole is becoming unsustainable and it does effect the value I get from Twitter.

    Can I ask what made you follow that 20% in the first instance? Did they genuinely interest you to begin with and now that has changed? Have your own reasons for following them changed? Or in some cases were you only following to be polite?

    Having had to face up to people’s obsession with “follower count as a metric” its good to see a healthy attitude towards what is often a completely empty number with no value.

    You’ve decided what’s valuable to you and it obviously wasn’t a number. Bravo sir.

    P.S I really wish you had checked your Klout score before and after your cull. For a service that claims not to put an emphasis on follower numbers it would have been interesting to see if your numbers plummeted or not. I’d do it myself but I’m too much of a chicken to potentially offend someone.

    1. Thanks Kelly. God only knows why I followed some of those people in the first place to be honest. Naivety? Politeness? Don’t know. Don’t remember. Some of them went back the best part of three years so it’s all a bit hazy. In some cases it was clear that there had been a valid reason to follow someone way back when but, on closer inspection it was evident that the noise to signal ratio for these people had dramatically increased over time. On reflection I think that was a fairly common factor in those decisions to unfollow that actually required more than a second’s thought.

      Hmmm, Klout. I’ve never visited the site. Never been tempted to score myself or others. There was enough informed commentary on Twitter and some of the blogs to which I subscribe to make it clear that it’s a bogus concept. I keep going back to this Slideshare presentation by Paul Adams (then of Google), which has some of the best analysis of how social networks work that I’ve ever seen. Real “influence” – if indeed influence has any bearing on what you use Twitter for – comes from strong ties, not a vast network of loose ties.

      And I’ve had enough experience with things like EdTwinge to know difficult it is to meaningfully analyse opinion and behaviour using only an algorithm, without overlaying any human interpretation of the data.

      Cheers. Phil

  4. Interesting post, Phil. Some great comments too.

    I’m increasingly asking myself why I use each of these different social services, including Twitter. What are the specific reasons to use each one, and how can I get the most value from it?

    I find it a huge challenge to follow more than around 150 Twitter accounts. (I’ve given up using ‘people’ as the noun for this, even though ‘accounts’ sounds slightly pompous.) I follow these accounts because I’m interested in what they have to say, so I like to keep track of everything they post. But I’ve only got a finite amount of attention.

    I think the danger is that Twitter just becomes a dumping ground for every thought, post, joke, quote, link, check-in and photo imaginable. I’d struggle with the signal to noise ratio if that trend continues.

    I wrote about this topic almost exactly a year ago. It’s a bit of a rambling post, and it’d come out differently (and shorter) if I wrote it today. But I don’t think my overall sentiments have changed a huge amount.

    1. Thanks Matthew. As our American cousins would say, we’re “parking in the same garage.” A flight to quality is most definitely where it’s at. The only real quantitative objective that I have left is to post two or three comments a week on other people’s blogs. I’ve more or less kept this up for the whole of 2011 and it is by some margin the most rewarding aspect of my social activities. So your comment here really is appreciated. Phil

  5. I do the same thing whenever my following count tops 1000. Somewhere along the way, I decided that 1000 was the max number of people I could follow and still have a reasonably managed incoming channel of meaningful, balanced information. Too much more than that and I start to feel like I’m missing important conversations.

    I’ve also become more judicious with whom I follow. I haven’t followed reciprocally in years, but now I don’t necessarily even follow some of my most frequent conversationalists back because their general updates (aside from our individual 2 replies) may not be relevant to me.

    Don’t ever be afraid to manage your information intake. Twitter’s a service, not a karma ration.

    1. Interesting that 1,000 is the magic number. It’s certainly the number that prompted my rethink. One of the main problems with information overload is that it restricts the time you have to do interesting stuff for yourself. And if you don’t do interesting stuff, it’s much harder to strike the right balance between consuming other people’s stuff and contributing your own. The former dominates at the expense of the latter. This blog post was the result of doing something and sharing the learning. It’s had a bigger reaction than any blog post that reflects on stuff that other people have done. I guess you’ll have found the same with posts like this – Thanks for the comment.

  6. Wonderful post highlighting the wide variety of personal choice afforded via social media.

    As you noted on Twitter, it would indeed be practically impossible for me to follow in your steps to manually consider each of the 108,352 Twitter accounts following me. However, the variance in our approaches to Twitter and the level of participation there flows both from our respective worldviews and objectives for being online.

    I live to meet, get to know, and hopefully inspire people. Thus my large following and, by comparison, intense participation online. The cool thing about Twitter is, any given individual can use it for what it is: a tool and a means to an end. How the tool will be utilized depends on that end.

    However, when it comes to corporations and large brands, they should follow back every human who follows them. Why? Well, if for no other reason than to give brand ADVOCATES a private way to say, “Pardon me, but your slip is showing” or, “Pardon me, but you have a massive stalk of broccoli in your teeth.”

    The problem is, many companies do not have a worldview that values people as human beings. People are statistics, ratios, variables in formulas.

    US Airways beautifully showcases a major brand’s inability to BE social in a world dominated by social beings. At least they’re honest, though. Check out their Twitter profile. They essentially say, “Go ahead and listen to us, but not only will we ignore you now, we have no plans to actually listen to you via social media.”

    Sure, you can use the email address US Airways gives, but the response you get will be as inhuman as their Twitter profile. (And, after spending 2 months and having 6 different phone conversations with US Airways brass, I can assure you their culture precludes actually listening to and hearing customers.)

    Bottom line: individuals can obviously use social tools any way they want to; corporations and major brands can really embarrass themselves, and show their true colors, quickly on social media. Not listening to customers is bad form no matter what media.

    1. Trey. Thanks for stopping by and leaving a considered and clearly heartfelt comment. Brands, particularly fmcg brands, focus on Customers (capital C) more they do on consumers (small c), despite spending millions each year to “target” the latter. Social spaces represent an opportunity to make consumers feel more like Customers. But the whole thing is quite scary for brand managers who are used to arms-length, one way relationships with the people that actually buy their stuff. The closest they get to dialogue is the other side of a two-way mirror during the occasional focus group. Nice meeting you. Phil

  7. PS. For what it’s worth, I do have to let my home feed fly by unattended except for specific times when I stop to explore. However, I do response to every @mention and every DM from a human being. Most DMs are spam, so it does take a little bit of work to find the real messages. Even with 108,000+ followers and 107,000+ followings, the data flow is manageable for me. But, everyone is different. I’m somewhat of a prospector. Prospectors know they’ll have to go through a lot of stuff that’s not gold to find the gold. One huge Gould-mine find was Scott Gould. What Scott wrote about the history of Like Minds, and then what he wrote in the incredibly generous post entitled “I Love Trey Pennington,” demonstrate the potential gold within the debris of social media.

  8. blackwatertown

    Interesting post and I like the Stop Following Me graphic at the top.
    I follow people for various reasons – for news updates, because I’ve had dealings with them professionally, for fun (Shit My Dad Says) or because it’s well nigh impossible to catch their attention any other way. There’s one person I follow who I found it was easier, quicker and more effective to tweet than to talk to, even when I was sitting beside him. (He’s a bit flighty in a genius kind of way.)

    1. Thanks for the comment. As a result of which I’ve actually found and followed you on Twitter. Unless it has been staring me in the face I’ve never seen a link to your Twitter profile on your blog. Are the two deliberately unconnected, despite sharing a common avatar? There’s a whole community of literary enthusiasts and publishers out there on Twitter. You don’t use it to “reach out” (sticks fingers down own throat) to potential publishers then?

      1. Ah the irony.
        The blog and the twitter developed completely separately. I’ve never linked them, though it’s not disastrous to be found out.
        That sounds very ungracious. Sorry.
        I occasionally plug the blog via twitter, but I’ve tended to use it for work-connected business, so the space between twitter and blog have allowed me space to be less discreet. A bit of wiggle room, deniability.
        I realise that this is deliberately going against the whole idea of everyone and everything being connected, but I think that is not always a good thing – can lead to getting fired, etc.
        I briefly had the twitter feed on a linkedin page, but decided professional connections didn’t all need instant notifications of every smart alecky trivia – as well as the profundities of course – that I spouted or retweeted forth.
        But time has passed, things have changed – so perhaps it’s time I reconsidered some convergence of the parallel lines.
        Hmmm, strange desire all of a suddenj to listen to Blondie.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.