This is pretty interesting. I'd love to see the analysis broken down by some measure of the ads' "annoyingness." For example, what is the harm from an animated gif ad vs. a text ad?
Gwern's Law: Ads cost you a tenth of your users
71–80 of 323 posts
Re: Gwern's Law: Ads cost you a tenth of your users
#72> So now there are 4 independent quasi-/experimental datapoints (me, Pandora, LinkedIn, & Mozilla resp) showing that a standard load of ads vs no ads decreases total consumption/activity by 10–15%. From the tweet itself, this is 10% of activity, not users. Kind of suspect that someone misinterpreted their own data so significantly.
I disagree about it being a misinterpretation. They are measuring different things because they have to, but if you read the papers, you see that that's the net effect and most reasonable way to summarize them: an 11% reduction in intention to visit, a 9% reduction in total traffic, a similar reduction in total web browsing per individual, and in the case of LinkedIn, you can see both a decrease in users and activity per user, which net to the >10% reduction in total activity. This is as it must be (what is a web browser user going to do - stop using web browsers completely and quit the Internet entirely...? Not likely! Reduction in web usage is their only possible response.) How else could one reasonably summarize that in a few words?
Re: Gwern's Law: Ads cost you a tenth of your users
#73I believe it. I've stopped playing games with ads altogether. I'd rather pay $40-60 for a game that keeps me in the story. I don't have time for ads. Apparently 90% of users do though, so they'll continue to fill up the "top games" lists and make it impossible for us to find good ad-free games in the app stores. Even worse now that google removed the "contains ads" tag from the play store top game lists. sigh
What bothers me is imagining what board room conversation led to that decision. Here's a piece of information, that benefits the consumer, and they took it away. It's hard to see much motivation beyond "it makes money, by selling more ads, because the user is uninformed". If you have a job in that position, yes I know money, but how the fuck do you justify that to yourself? Are you really that blind?
Re: Gwern's Law: Ads cost you a tenth of your users
#74If a game doesn't offer any way to pay to remove ads, it's an instant uninstall for me. If a game does offer it, but doesn't let me play for long enough to decide whether I like the game before it starts serving up unskippable ads, I uninstall it. Most games simply don't give enough time to decide if you even like it before they start throwing ads at you. That seems like a terrible policy in general, regardless of wh…
I’m about to release a fun little ambient physics based puzzle game on iOS and have been debating this. To me ads are a stain on what I consider my art. But at the same time I want to monetize and wonder if people will skip over a $1.99 download. What do you think? Ads with the option to unlock? Seems gross though... maybe free with paid version that includes level editor or additional level packs?
Free features to get the user to get used to the game and figure it out, and paid features that are unlocked.
App pricing is somewhat tricky, since most people hesitate to even pay $0.99. I personally wouldn’t just buy something without knowing if I’d like it or not. Descriptions and reviews may not always help in knowing that. Actually playing it certainly would.
Also consider keeping some exciting or challenging parts in the free set so that users can know what more to expect if they pay.
Re: Gwern's Law: Ads cost you a tenth of your users
#75If a game doesn't offer any way to pay to remove ads, it's an instant uninstall for me. If a game does offer it, but doesn't let me play for long enough to decide whether I like the game before it starts serving up unskippable ads, I uninstall it. Most games simply don't give enough time to decide if you even like it before they start throwing ads at you. That seems like a terrible policy in general, regardless of wh…
I’m about to release a fun little ambient physics based puzzle game on iOS and have been debating this. To me ads are a stain on what I consider my art. But at the same time I want to monetize and wonder if people will skip over a $1.99 download. What do you think? Ads with the option to unlock? Seems gross though... maybe free with paid version that includes level editor or additional level packs?
For a productivity app I have seen night and day difference in earnings between "paid upfront" and "free with iap". I started with the paid model and switched to freemium. In the months after the initial boost due to "apps gone free" scrapper sites I got as many iap sales in a month that I had accumulated in two years while the app was paid.
Paid upfront model is dead on iOS, unless you are some big shot company that will have their application featured on day 1.
Re: Gwern's Law: Ads cost you a tenth of your users
#76If a game doesn't offer any way to pay to remove ads, it's an instant uninstall for me. If a game does offer it, but doesn't let me play for long enough to decide whether I like the game before it starts serving up unskippable ads, I uninstall it. Most games simply don't give enough time to decide if you even like it before they start throwing ads at you. That seems like a terrible policy in general, regardless of wh…
I’m about to release a fun little ambient physics based puzzle game on iOS and have been debating this. To me ads are a stain on what I consider my art. But at the same time I want to monetize and wonder if people will skip over a $1.99 download. What do you think? Ads with the option to unlock? Seems gross though... maybe free with paid version that includes level editor or additional level packs?
For what it's worth, the only Android games I've ever paid money for are Snakebird, Ending, Super Hexagon, and a chess training app, and if I remember correctly all four were free no ads with paid additional levels. I am the type of person who deletes apps with too many ads. Look up Ending for a good model to emulate (it was a pleasant experience and it has a level editor) though I remember it having too few levels since I played through them very quickly.
Re: Gwern's Law: Ads cost you a tenth of your users
#77I have been seeing Squarespace ads on YouTube nearly for every video. I hate it and I just gives thumbs down hoping it will go away. Nope.
Although the echo effect is real, I have to say I do not and will not use Squarespace simply because of the ads. It's insane. Same music, same video over and over.
/rant
Re: Gwern's Law: Ads cost you a tenth of your users
#78Earlier quoted context omitted.
Are we controlling for SEO effect? Having ads on your site probably nerfs your Google ranking a bit.
Wouldn't the incentives go the other way? I'd imagine Google wants people to get to sites that serve ads from Google so they can collect fees.
Ads don't affect SEO. It would allow websites to essentially pay for higher search results.
There are tools that come with using ads that allow for insights into better SEO though - e.g. Adwords keyword planner.
Re: Gwern's Law: Ads cost you a tenth of your users
#79Additionally, 26 percent of desktop and 15 percent of mobile users block ads, so those studies already exclude the most ad-averse part of the user base.
Re: Gwern's Law: Ads cost you a tenth of your users
#80> So now there are 4 independent quasi-/experimental datapoints (me, Pandora, LinkedIn, & Mozilla resp) showing that a standard load of ads vs no ads decreases total consumption/activity by 10–15%. From the tweet itself, this is 10% of activity, not users. Kind of suspect that someone misinterpreted their own data so significantly.
I disagree about it being a misinterpretation. They are measuring different things because they have to, but if you read the papers, you see that that's the net effect and most reasonable way to summarize them: an 11% reduction in intention to visit, a 9% reduction in total traffic, a similar reduction in total web browsing per individual, and in the case of LinkedIn, you can see both a decrease in users and activity…
In analytics there are ways to discuss activity like: visits per month, pages per session, time on site, conversion rate, etc.
It's possible for any of those activity metrics to go down 10% without losing any active users.