On Wed, Sep 22, 2010 at 1:27 AM, Andre Engels <andreengels(a)gmail.com> wrote:
On Wed, Sep 22, 2010 at 9:55 AM, Lennart
Did you know that less than a third of the users
who create an account on
English Wikipedia make even *one* edit afterwards? Two-thirds of all new
accounts never edit! Interestingly, this percentage vary very much from
language version to language version.
Now, the question is not: "what can we do about it?" We know plenty of
things that we *could* do. The question is this: "what are the easiest
levers to push that increase the numbers?"
I think we need to take a step back first. Before deciding on what to
do about this, two other questions have to be asked:
1. Why are people creating an account without editing?
2. Do we want/need to do something about it?
There are various reasons why people could register without editing.
To name a few:
* people coming in from other Wikimedia wikis, auto-registering through SUL
* people creating sock puppets, then not needing them or forgetting
them when they do need them or having them blocked before they have
the chance to use them
* people wanting to change personal settings
* people thinking they can get something extra by registering
(previous category, but then for personal settings that don't exist)
* people who out of a habit register for every web site they see where they can
* people who find they cannot create an article on en: wikipedia
unregistered, want to create an article, register and then find
creating an article is too difficult
Not all of these categories we want to do something about their
non-editing, and when we do want to do something about it, we should
not use the same strategy on all. Therefore, before trying to solve
the problem, I think you should first determine
1. whether there is a problem, and
2. if so, what the problem is
Some recycled stats for EN wikipedia from many months ago:
~ 215 k account creations / month (~15% imported from other wikis)
~ 65 k new accounts will edit at least once
~ 22 k new accounts will edit at least 5 times
~ 8 k new accounts will edit at least 10 times
~ 2.4 k new accounts will eventually edit at least 100 times.
There is a huge gap between the number of registrations and number of
new accounts that edit. However, there is also a huge gap between the
number of people who edit at least once and the number that edit at
least 10 times.
From the user interaction tests done by the Usability
is clear that a significant portion of visitors are confused,
intimidated, scared, or otherwise discouraged when they first
encounter the editing process. I don't think it takes a big leap to
conclude that feeling uncomfortable with making edits is a major
reason that people abandon new and lightly used accounts (though it is
certainly not the only reason).
Currently, only about 1.1% of account registrations reach the level of
100 edits. That conversion rate is tiny. Even if there are many
different reasons that such people abandon editing, I have to imagine
that a significant portion of the other 99% are people we can capture
and develop into active editors given a bit more support. Even if
such efforts can only convince another 1% to become active editors,
you would still be talking about doubling the size of the active
So I certainly support initiatives that reach out to and help the very
newest users. Exactly what kinds of support we provide will likely
vary depending on what issues are determined to be most common, but
regardless of the details, I think this is a very important avenue for
outreach and growth.