Popular on Pinterest? I spoke too soon...

Now here's a funny thing: my last post was about how the number of followers for my experimental Pinterest board, Places as They Used to Be had suddenly taken off, rising to six thousand in a couple of weeks, after taking a couple of months to reach the first thousand.

My guess was that the board was one of a selection being suggested to new joiners as something to follow to get them started on Pinterest.

But from almost the exact day the blog post was published, the phenomenon came to an equally sudden end. In the few days since, new followers have slowed to a trickle (and even fell on one day). Have a look:

Could it be that the Pinterest staffer I tweeted about it, having decided not to reply to me, also decided to end my place in the new-joiner limelight? 

That's probably paranoia - I expect they've got better things to do. But I'll be interested to see what happens next.

Other thoughts about Pinterest - by me and others - are gathered here. 

The Pinterest files

Here's a collection of stuff I've written about Pinterest, starting with the most recent:

1. Popular? I spoke too soon.

2. Why I'm suddenly so popular on Pinterest (June 18 2012)

3. What does Pinterest users' favourite colour say about them? (March 4 2012)

4. Getting hooked on Pinterest (Feb 20 2012)

5. For the BBC College of Journalism: Test driving the latest self-expression engine (Feb 10 2012)

And here are a few other bits and pieces about it:

- Mark Zuckerberg's Pinterest account

- Pinreach measures your Pinterest clout. It offers the exciting news that my Pinterest account (second from left, below) has a higher score than NBC News:

- Influential blog post from Business Insider about copyright fears re pinning

- More recent blog warning of copyright issues re Pinterest, with some useful references

- Pinterest's own blog

- Report on interview with Pinterest co-founder Ben Silbermann

- If you have an answer to my question "How come when I start a Pinterest board it immediately has 600 followers?", which I posted on Quora, please add it here, because nobody else has at the time of writing... Thanks!

- Finally, in case you haven't come here from it, this is my Pinterest home page.

PS: If you are interested in Pinterest experiments, you might also be interested in my Twitter experiments, which are somewhat like this: read about my attempts to build a Twitter empire with no real work are here. 

Taking off on Pinterest

I have already explained how my attempts to build an audience on Twitter automatically were unspectacular but there's better news from Pinterest.

Look what's been happening to follower numbers for my Places as they used to be board in the past few days:


After taking about three months to reach 1000 followers, the board has grown to 6000 in a little over a fortnight. And I haven't been doing anything different.

What's going on? I'm guessing that my name or the name of that board is now being offered to people when they sign up to Pinterest.

That looks right because many of the recent followers (below) haven't themselves got followers: in other words, they've just joined:

Compare that with a sample of earlier followers, who look like old Pinterest hands:

This lot probably started following my board in the normal way - seeing a picture from it on the Pinterest home page and clicking to follow the board. 

But there are other mysteries to Pinterest's numbers: either I don't understand how they work or they don't work:

So, for instance, why don't the follower numbers of a board equal the person's follower numbers plus the number who have followed that particular board?

That isn't the case: apparently 200 more people follow me than follow any of my boards. By the above logic, the board numbers should be higher than the person numbers.

Then there's this: when I set up a new board, it immediately had 624 new followers. And these aren't just the people who follow me, because that number is around six thousand. 

I came across this phenomenon because I set up a board with just patterns to test whether I'd even get followers for that. Indeed I did: to my surprise, 624 instantly (up to 667 a couple of days later). 

So what about an empty board I set up to test an even more extreme case? Again, about the same number of followers from the moment it was set up (and 668 now). See below (well I put one pic on it now).

Whatever's going on here, I think we can take it that several hundred people don't actually want to follow an empty board: Pinterest is somehow massaging the figures to provide encouragement for its users and make them think there's more interaction than there really is. 

Although I'm a bit surprised by my sudden increase in followers, the numbers are peanuts compared to some boards. 

Take Jane Wang of Chinstrap Penguin, Antartica: more than 3 million followers for her 89 boards (e.g. Happy - 2.2m followers).

How did that happen? She's not Stephen Fry or Lord Sugar, yet she's got their kind of Twitter clout on Pinterest, and there are plenty of others in the same league. 

Are there penguins on Pinterest?

Richmond Park after rain

On July 28 and 29, the Men's and Women's Olympic Road Cycling Race will be going through - but I don't suppose the riders will be enjoying the views. 

Going local automatically on Twitter: an experiment

Just over a year ago I set up a collection of Twitter accounts, fed automatically with tweets on different subjects from a news aggregator. As I wrote here, most of them haven't exactly taken off.

But if there was a lesson from the experiment, it's that local-focussed accounts are a better bet than other subjects. The most popular of my collection is East Sheen News (below), which has 123 followers. Not many, but actually more than any other East Sheen account - even those written by real people rather than fed automatically.

Encouraged by that, I also started, for the same locality, Barnes News (52 followers) and Mortlake News (18 followers). A collection of geographically-related accounts gives me the advantage of being able to retweet between them as a kind of cross-promotion.

Their follower numbers are in proportion to the length of time time the accounts have existed, the number of tweets and the amount of attention I've given them (I occasionally tweet 'real' tweets to supplement the automated ones).

If you want to start a local news site or Twitter feed with automated content, it helps to pick somewhere whose name is only the name of the place: so setting up a search for news stories with "East Sheen" in will almost always produce the right kind of results, whereas "Barnes" can produce stories about the footballer, or other things that aren't the Barnes you're looking for.

Inevitably stories do appear which have nothing to do with the area: for instance, Richmond Park is another of my search terms for these accounts, and it turns out there's a Richmond Park in Norfolk, so stories about that pop up. Sometimes I get tweets from clever clogs accusing me of being an automated feed. I like to reply with a gracious apology, admitting that it's all automated and, well, you know what those automated searches are like...

What's fun about this kind of account is acquiring an spurious reputation as a local opinion-former. Several local businesses, such as pubs and restaurants, have asked me to retweet something they're trying to promote. I even got invited to the opening of a local restaurant.

I've never said who I am in my accounts, and nobody has ever asked. Which makes me wonder if I could just as easily start one for, say, a small town overseas. Could I become an opinion-former somewhere thousands of miles away, maybe somewhere I've never been?

Well, as I said, none of these accounts exactly hit the jackpot.

Another idea I had to create a network of interest was to pick an event some months hence and set up a Twitter account, thinking it might pick up momentum as the event came into the media spotlight.

So last year I set up FacebookIPOnews and LD2012_today (about the Olympics).

Again, disappointing: the Olympics picked up just 40 followers in a year (below) - but, like the other accounts, with no work from me except the 15 minutes or so it took to set up on day one:

My second event-related account, FacebookIPOnews, picked up just 8 followers, which is extraordinarily poor: I've seen accounts which have never tweeted that have more followers that than. This is real mystery: FacebookIPOnews has been churning out pretty relevant news links (1903 in all), to no effect. None of my other accounts has anything as low as that kind of number: I just don't know what the explanation is. My Facebook News, for instance, has 55 followers - with a similar amount of non-effort.

So overall, my Twitter experiment was interesting but didn't uncover a way to achieve influence without putting in the hours.

The only place where I've succeeded in building a more substantial audience with very little work is on Pinterest, where my Places As They Used To Be board has built 2356 followers in much less time than the above accounts. More about the curious process behind that next time.