12 ms·
CSS changes that increased Reddit account creation by 200K yearly users
- theVirginian 12y agoIMO the design of Reddit's website could use an overhaul. I can't even snap it to one side of my screen without totally screwing up the front page and forget about using the site on mobile. The sidebars contain way too much unnecessary information as well. I do love me some reddit though.
- baddox 12y agoIt's refreshing to see what is to my eyes clearly a cleaner and more modern design actually performing better than what looks like a kludgy FrontPage design. Too often these analyses purport to show the opposite.
- herbig 12y agoI'd say they're correlated for sure.
- brent_noorda 12y agoMy sarcasm detector is having trouble. Was your comment sarcasm? Was the original article and accompanying graph sarcasm? Is it sarcasm to say that unix time is incrementing at the same rate in 2015 as it did in 2014, and therefore it being 2015 is responsible for the increase in unix time?
- herbig 12y agoNo. As many of the other comments have here have said in more words, correlation != causation.
- gggggggg 12y agothough, doesn't this suggest Unix time will not increment at the same rate throughout the year. http://hpiers.obspm.fr/iers/bul/bulc/bulletinc.dat http://hpiers.obspm.fr/iers/bul/bulc/bulletinc.dat
- AdamGibbins 12y agoThey removed the CAPTCHA, I'm not surprised.
- extrapolate 12y agoQuote from a previous article (so I wouldn't attribute all 8% to bots): We’re experimentally removing CAPTCHA and using other methods for spambot detection. We know it’s annoying for humans, and are starting to worry the bots are actually enjoying it. Plus, our visual CAPTCHA was failing on accessibility for non-visual redditors.
- AdamGibbins 12y agoSorry, to be clear, I wasn't attributing the lack of CAPTCHA to bots. Rather the improved usability and increased human sign-up rate to the removal of a painful CAPTCHA.
- jbeja 12y agoWhat other method for spambots could be used?
- bobbles 12y agoIn addition to plenty of stuff I cant think of, you could see: 1) High percentage posts flagged as 'reported' 2) High percentage posts flagged as 'spam' 3) High volume of posts containing URLs 4) Posts containing identified spam messages (like how an email spam filter may work) 5) Multiple accounts from similar/same IPs Almost all spam in a subreddit I mod for is from accounts < week old. (I believe there may already be a limit on how much you can post after registration though)
- ubernostrum 12y agoI mod a subreddit with >100k subscribers. We use AutoModerator and have it set to remove anything posted by an account less than a day old. It catches some false positives from people who sign up just to post something, but it's also drastically decreased the spam. Most spambots are still at the "sign up and then blast shotgun-loads of posts before the algorithms catch you" level of behavior.
- xpose2000 12y agoI wish I had a screenshot of the original version. It was absolutely awful. It seemed like the code wasn't touched in at least 5 years. Nice job with the redesigned version, but in all seriousness... I wonder what took so long? Every few months a company should take a spin as a new user and find ways to refine the new-user process. Use this as a reminder to take another look at your registration sign-up process in 2015.
- peeters 12y agoThere's a link to the original article about the redesign included in this article. It has a comparison of screenshots. Here you go: http://www.donotlick.com/2014/11/17/redesign-of-reddits-loginaccount-creation-window-and-reddit-comlogin/ http://www.donotlick.com/2014/11/17/redesign-of-reddits-logi...
- fixie 12y agoThere is a screenshot shown in the linked previous article: http://www.donotlick.com/2014/11/17/redesign-of-reddits-loginaccount-creation-window-and-reddit-comlogin/ http://www.donotlick.com/2014/11/17/redesign-of-reddits-logi... Direct link: http://www.donotlick.com/wp-content/uploads/2014/11/cheat.png http://www.donotlick.com/wp-content/uploads/2014/11/cheat.pn...
- tuna 12y agoany way to check which ones are real people and which ones are bots creating accounts for some reason ? maybe their activity posting vs subreddit popularity... I guess you may be surprised.
- xiongchiamiov 12y agoWe ran some analyses and determined the rise is not due to an increase in spambots. I don't know the specifics of how that was determined, sorry.
- Fogest 12y agoHow does this actually indicate it had to do with the redesign? Sure the trend may be going up, but the site is increasing in popularity so that makes sense that it is going up. There are more people there to interact with and more communities people can join that they may be interested in following and discussing things in. I am just wondering how this rise is attributed to the CSS change? It may have had some effect, but I'm confused how this single graph says that it is because of a CSS change?
- richardwhiuk 12y agoIn fact, from my reading the rate of signups was increasing before hand and continued increasing afterwards, implying the design made no difference. That's what I can see from the trendline anyway, and the rest of the graph is too uneven to draw any conclusions.
- Fogest 12y agoYes, exactly. The graph is full of spikes and I do not see any indication that there was any real change present from the CSS change.
- eCa 12y agoThey are wrong to draw the trendline the way they did. Since what they are claiming is not steady growth (then the trendline would make sense) but a shift in behaviour they should instead have drawn one line for average before the change, and one line for average after the change. It is quite obvious that something has changed. Before the change there are five days above 7.5%, and after there are more than 10 days above 7.5%. The valleys are much higher too.
- cbsmith 12y ago> It is quite obvious that something has changed. Yup, but that's not an 8% increase in AR's. That's maybe an 8% increase in the rate of increase in AR's. That could be attributable to a LOT of things though (there isn't even enough context to control for network & seasonal effects, let alone whatever increased attention there might have been to sign ups after the rollout). Ironically, probably the best way to isolate the impact to the launch is to narrow the time window and resolution to the minutes before and after the redesign launched.
- TomGullen 12y agoSorry but this chart: http://www.donotlick.com/wp-content/uploads/2015/01/newdesign.png http://www.donotlick.com/wp-content/uploads/2015/01/newdesig... Doesn't seem convincing evidence to me
- rcfox 12y agoI agree. While it does show an average increase of ~7.5% after the line, it also showed that before the line! It's hard to tell that anything happened.
- jbeja 12y agoYeah, that what a properly modern designed account form given today standards can do this days.
- pimlottc 12y agoSubmission title is a bit clickbaity and does not match the more modest actual article title, which is "8% Increase in reddit Account Registrations".
- supercoder 12y agoJust imagine what would happen if they applied better design to the rest of the site !
- megablast 12y agoThe subs have a lot of control over what is presented. I think they often look worse, but that is all a design choice.
- JungleGymSam 12y agoDesign is a choice? Radical thinking.
- toddmorey 12y agoOne quibble: there are way more than 'CSS' changes here. They removed extra wording, removed the CAPTCHA, and generally reorganized the window.
- megablast 12y agoWell, if there were lazy they could have just hidden all those other elements.
- sideproject 12y agoA bit more data would be nice to corroborate the claim. I understand some of the criticism in the comments. :) If you are looking for some ridiculous correlation claims, check out the spurious correlation. Here's an example, The number of people who drowned by falling into a swimming pool correlates with the number of films Nicolas Cage appeared in. http://www.tylervigen.com/view_correlation?id=359 http://www.tylervigen.com/view_correlation?id=359 :)
- underwater 12y agoAnd yet it is still impossible to use on a mobile device, due to a crappy custom dialog implementation.
- underwater 12y agoAnd yet it is still impossible to use on a mobile device, due to a crappy custom dialog implementation.
- vii 12y agoThis is a typically poor example of ignoring the complexities inherent in analysing new user acquisition. Suppose you implement a change that causes the registration form to crash for some subset of people? I am sorry to say for a while I was responsible for causing such issues at a very large service ;) Depending where the crash occurs and where you are counting signups you may actually get a massive boost in your metrics, as the victims retry with every attempt met by failure. Suppose you correct an issue making it easier for people to login? Then lower registrations - the key here is to make sure you have a good metric to look at (e.g. accounts created that are interacting well with the site after as long a window as possible).
- JungleGymSam 12y agoThe article is about an 8% increase in account registrations not hits.
- vii 12y agoExactly. An account registration is basically like a hit metric. The creation of an account does not imply that it will be used or the person who created it has not created many more either accidentally or for example because they could not log in.
- micheljansen 12y agoShame that they didn't do some A/B testing to more accurately quantify the effect.
- mahouse 12y agoThey probably have the same answer they have for the "Why can't I search comments" question: they can't afford it.
- outsidetheparty 12y agoThat chart is a textbook example of "we drew a trendline through noisy data and drew completely unwarranted conclusions as a result".
- chestnut-tree 12y agoThe before-and-after image shows a much less cluttered sign-up form http://www.donotlick.com/wp-content/uploads/2014/11/cheat.png http://www.donotlick.com/wp-content/uploads/2014/11/cheat.pn... However, the Nielsen Norman Group (NNG) recommend against using placeholder text in form fields. They list 7 main reasons. This is taken from their website 1. Disappearing placeholder text strains users' short-term memory 2. Without labels, users cannot check their work before submitting a form 3. When error messages occur, people don’t know how to fix the problem 4. Placeholder text that disappears when the cursor is placed in a form field is irritating for users navigating with the keyboard 5. Fields with stuff in them are less noticeable. 6. Users may mistake a placeholder for data that was automatically filled in 7. Occasionally users have to delete placeholder text manually Obviously, not all of these will apply to the Reddit sign-up form (or to your own website). It's worth reading the full NNG article to get a different but valid perspective based on their research: http://www.nngroup.com/articles/form-design-placeholders/ http://www.nngroup.com/articles/form-design-placeholders/