Redesigning "Add-ons"

[mikelove]

We're starting to sketch out designs for a new "Add-ons" screen (probably for the update after the next big one), a vitally important part of Pleco which we've neglected for too long: the current design is quite poor, particularly for its primary purpose of "convincing people to buy stuff," but (much like our also-neglected website) it's something that nobody outside of Pleco is particularly upset about, so it's had a hard time getting to the front of our task list.

So: what sorts of information would you find helpful when deciding whether or not to purchase a particular add-on? My inclination is to kill the "demo" idea for dictionaries altogether - kind of a Palm/WM-era relic now that we ship our software with a couple of complete free dictionaries - but then provide a lot more info on the add-on page for each dictionary: a couple of paragraphs of description, a bunch of easily-viewable sample entries, perhaps some user reviews, etc. Non-dictionary add-ons might include video walkthroughs or somesuch, though we'd probably also keep as many of them as we can demo-able (handwriting and OCR are definitely things that are best understood / appreciated through use).

 

[character]

I think you make a good separation between dictionaries and other add-ons, which I'll call features. For features, I think you should provide more information about how they enhance each other, so people know OCR has a way to quickly create flashcards, etc etc.

Video walkthroughs would be great, if not in the app for size reasons then somewhere online. I suggest showing scenarios of users using the features together to enhance their learning efficiency.

For dictionaries, what you describe sounds good to interest people.

I don't think it's against the rules to do A/B testing with add-on descriptions in the app and see what works best.

Goodreader has a "what's new in this version" popup which appears the first time you launch it after an update. You could do the same, and mention any new enhancements to add-ons.

I've not seen the free version in quite a while, so you may be already doing this sort of thing: I don't think people would begrudge you being a bit more in your face with pushing add-ons in Pleco if it becomes unobtrusive once someone has purchased at least one add-on.

 

[Alexis]

An older comment I had from an Android thread:

Also, I was a bit puzzled by the meaning of "non-dict" (at least I assume the meaning of this would be puzzling for the new user) in the Bundle descriptions. Maybe a better name might be "Extensions" for the non-dictionary add-ons. In such a case, it may be clearer to group the add-ons in each tab under categories (ie. Bundles, Dictionaries, Extensions).

Edit: Also, I don't see a need to list free/paid separately.

 

[scykei]

Video walkthroughs would be great, if not in the app for size reasons then somewhere online. I suggest showing scenarios of users using the features together to enhance their learning efficiency.

I disagree. I for one, hate watching video walkthroughs. I'm not going to spend my time on that if I can help it. It would even prefer a help in text than video because it is easier to jump to whichever relevant parts that I want to read. If it were a video, I have to watch everything whether I like it or not before I can find what I want.

Videos that demonstrate what it can do overall is fine for advertising, but not video walkthroughs.

I've not seen the free version in quite a while, so you may be already doing this sort of thing: I don't think people would begrudge you being a bit more in your face with pushing add-ons in Pleco if it becomes unobtrusive once someone has purchased at least one add-on.

What does that mean? Annoy users until they make their purchase? I don't like that idea. The best way to convince someone to buy it is if they are absolutely happy with what they are using and would want to go for more. Advertisements, pop-ups and other annoyances will only dampen their interest in it.

I like what Mike has planned, however. Better and more descriptive information about the add-ons would be very useful. Right now, there isn't really anything convincing to persuade someone to buy the add ons.

 

[Alexis]

Maybe a more subtle way of pushing addons is to internally use a a Number badge on the Add-ons icon (only visible when the user presses the fan/menu icon). The number in the badge would be the number of new/updated addons, and would go away as soon as the add-ons screen is visited. The badge would only return when there is a new add-on and/or a purchased/downloaded add-on has been updated.

 

[character]

I disagree. I for one, hate watching video walkthroughs. I'm not going to spend my time on that if I can help it. It would even prefer a help in text than video because it is easier to jump to whichever relevant parts that I want to read. If it were a video, I have to watch everything whether I like it or not before I can find what I want.

Videos that demonstrate what it can do overall is fine for advertising, but not video walkthroughs.

Perhaps it's just a disagreement over the word 'walkthrough'? Mike suggested he was considering video walkthroughs as a sales device, not help. Pleco.com has text help already, and there is a downloadable manual for the app.

What does that mean? Annoy users until they make their purchase? I don't like that idea.

I don't mean annoy users, but there's nothing wrong with (assuming these would work, not having seen the free version in a while):

1. User taps the dictionary switch button (but only has the Pleco dictionary installed) and gets a page explaining that other dictionaries are available to purchase on the add-on page, and information about the various dictionaries available. If the user purchases a dictionary, then that upsell page doesn't appear when the button is pressed.

2. User taps the dictionary search box and sees the segmented control for Rad | Key | HWR. If the user clicks on HWR, a page appears explaining Pleco's advanced handwriting recognition is available for purchase, etc. etc. as above.

I trust that Mike could work in such information about add-ons available for purchase in a way which was not annoying.

The best way to convince someone to buy it is if they are absolutely happy with what they are using and would want to go for more.

Unfortunately, app store developers have found if people are happy with what the free version provides and would have to look to find things to purchase, very few people do purchase enhancements to apps.

Advertisements, pop-ups and other annoyances will only dampen their interest in it.

That would explain the massive failure of freemium games and apps on the app store last year. http://gigaom.com/mobile/freemium-app-r ... -the-dust/

Better and more descriptive information about the add-ons would be very useful. Right now, there isn't really anything convincing to persuade someone to buy the add ons.

I agree. But it's not clear from Mike's post if people are regularly visiting the add-on page and then not purchasing any add-ons. If enough people aren't viewing the add-on page, then something needs to prompt them to do so.

 

[scykei]

What does that mean? Annoy users until they make their purchase? I don't like that idea.

I don't mean annoy users, but there's nothing wrong with (assuming these would work, not having seen the free version in a while):

1. User taps the dictionary switch button (but only has the Pleco dictionary installed) and gets a page explaining that other dictionaries are available to purchase on the add-on page, and information about the various dictionaries available. If the user purchases a dictionary, then that upsell page doesn't appear when the button is pressed.

2. User taps the dictionary search box and sees the segmented control for Rad | Key | HWR. If the user clicks on HWR, a page appears explaining Pleco's advanced handwriting recognition is available for purchase, etc. etc. as above.

I trust that Mike could work in such information about add-ons available for purchase in a way which was not annoying.

No. I don't like that idea. I have mentioned this before somewhere. Adding extra buttons that will not work feels very cheap. I love the way Mike handles everything right now. The audio buttons, handwriting interface, OCR, flashcards, etc have their buttons hidden until the user installs the demo, which will be done willingly. This makes the app feel complete instead of merely a trial software that wants your money.

The best way to convince someone to buy it is if they are absolutely happy with what they are using and would want to go for more.

Unfortunately, app store developers have found if people are happy with what the free version provides and would have to look to find things to purchase, very few people do purchase enhancements to apps.

That is one of the worst ideology when designing a lite app. Have you heard of the app called Trainyard? It has a lite version that most consider to be the right way to do things.

The devs made a blog entry that I find very inspiring. Check it out here:
http://struct.ca/2010/the-story-so-far/

Advertisements, pop-ups and other annoyances will only dampen their interest in it.

That would explain the massive failure of freemium games and apps on the app store last year. http://gigaom.com/mobile/freemium-app-r ... -the-dust/

I have no idea if you're being sarcastic or not. A fremium game is usually the apps that have everything free, but provides in-app purchases to use real money to gain advantage in game. And if you look at it, none of the freemium games that are at least a little bit successful have disgusting pop-ups and advertisements prompting users to buy stuff. They will draw you into the experience, finally causing users to cough up money due to the addiction. It's not a very ethical way as it ruins the gaming industry, but it's good for quick cash.

Watch this video by Penny Arcade to understand it:
http://www.penny-arcade.com/patv/episod ... kinner-box

Better and more descriptive information about the add-ons would be very useful. Right now, there isn't really anything convincing to persuade someone to buy the add ons.

I agree. But it's not clear from Mike's post if people are regularly visiting the add-on page and then not purchasing any add-ons. If enough people aren't viewing the add-on page, then something needs to prompt them to do so.

It will happen over time. Don't take away things that people already like. You don't want this amazing app to have people start complaining that they want the old version back, or saying they are "lucky to have saved a backup of the old version of the app", do you?

 

[character]

That is one of the worst ideology when designing a lite app. Have you heard of the app called Trainyard? It has a lite version that most consider to be the right way to do things.

The devs made a blog entry that I find very inspiring. Check it out here:
http://struct.ca/2010/the-story-so-far/

And in October 2011 he wrote a post entitled "In Defence of Freemium". Keep in mind that Trainyard and Pleco are very different animals. There's a free and a paid version of Trainyard, and the app store is a very different beast now than it was in 2010. I follow Matt Rix and a bunch of other iOS developers on Twitter. I don't claim to speak for them, but I don't think their views on the current app store are the same as yours.

And if you look at it, none of the freemium games that are at least a little bit successful have disgusting pop-ups and advertisements prompting users to buy stuff.

I suggest you start listening to Walled Garden Weekly, or at least the CSR Racing episode: http://media.walledgardenweekly.com.s3. ... /wgw-2.mp3

Watch this video by Penny Arcade to understand it:
http://www.penny-arcade.com/patv/episod ... kinner-box

Yes, I've seen all the Extra Credits episodes.

It will happen over time. Don't take away things that people already like. You don't want this amazing app to have people start complaining that they want the old version back, or saying they are "lucky to have saved a backup of the old version of the app", do you?

I guess we'll just have to agree to disagree; I don't see that what I've suggested is horrible. The free version of Pleco exists to sell add-ons; it's a not a gift to the world from a philanthropist. I want Pleco the company to make enough money to remain viable and keep developing Pleco app improvements.

 

[scykei]

Haha. Of course, I want the company to make money too.

I thought about it for a bit, and the point is, Pleco is a very expensive app by the Appstore's standards. We can't just do stuff like that to the users and expect them to buy the addons. What we need to do is gain their trust. There will be users that cannot afford to make that kind of purchase or just wouldn't consider to make any purchases above their normal budget. We can't force this group of people. If it were a just 99 cent app, then it would be a mindless purchase and there is a chance that this brute force method might work.

People like me, however, put a lot of thought into it before buying, especially something of this range. If you don't believe me, you can ask Mike how many tiresomely long emails I've sent to him regarding the purchase .

 

[mikelove]

Thanks for all of this feedback.

I think you make a good separation between dictionaries and other add-ons, which I'll call features. For features, I think you should provide more information about how they enhance each other, so people know OCR has a way to quickly create flashcards, etc etc.

Makes sense - we're actually hoping to highlight that distinction even more in the future, as we're now planning to start selling dictionaries outside of our app as well (since it seems to be within the rules and we'd very much like people to have a way to spend $100 on new dictionaries without having $30 of it go to Apple) and hence to push people towards buying their 'features' in-app and their dictionaries on our website.

I don't think it's against the rules to do A/B testing with add-on descriptions in the app and see what works best.

That I'd have to double-check on - Apple actually requires us to write up an official description of each IAP that goes in their database and gets reviewed / approved like any other bit of App Store text, but it's not clear what the requirements are concerning the display of that versus other text that we might come up with instead.

Goodreader has a "what's new in this version" popup which appears the first time you launch it after an update. You could do the same, and mention any new enhancements to add-ons.

Very logical, yes.

Also, I was a bit puzzled by the meaning of "non-dict" (at least I assume the meaning of this would be puzzling for the new user) in the Bundle descriptions. Maybe a better name might be "Extensions" for the non-dictionary add-ons. In such a case, it may be clearer to group the add-ons in each tab under categories (ie. Bundles, Dictionaries, Extensions).

Yes, "non-dict add-ons" is a terrible turn of phrase.

Edit: Also, I don't see a need to list free/paid separately.

I believe we did that one originally because we were trying really hard to get people to download CC-CEDICT on iOS too (it's optional on iOS, unlike on Android, largely because we wanted to keep our basic app under what at the time was a 20MB limit on download size). But now that we're planning to build in CC along with PLC it might indeed make sense to bury / categorize it.

I disagree. I for one, hate watching video walkthroughs. I'm not going to spend my time on that if I can help it. It would even prefer a help in text than video because it is easier to jump to whichever relevant parts that I want to read. If it were a video, I have to watch everything whether I like it or not before I can find what I want.

Videos that demonstrate what it can do overall is fine for advertising, but not video walkthroughs.

I don't care for them either, but a lot of people seem to find them far preferable to reading, so I think there's something to be said for both approaches - come up with a script and then simultaneously offer it in video and screenshots-with-text form. Videos make sense for advertising too - rapidly go through the features to make them ooh and aah and overwhelm them with the sheer quantity of stuff we have on offer - but making detailed instructional videos available might actually help in persuading people to hit "buy."

Maybe a more subtle way of pushing addons is to internally use a a Number badge on the Add-ons icon (only visible when the user presses the fan/menu icon). The number in the badge would be the number of new/updated addons, and would go away as soon as the add-ons screen is visited. The badge would only return when there is a new add-on and/or a purchased/downloaded add-on has been updated.

Possibly, but in other apps I've found that if anything that has the opposite of the desired effect - I go to that screen with the single-minded purpose of getting rid of the annoying red dot and don't bother reading what it has to say.

It will happen over time. Don't take away things that people already like. You don't want this amazing app to have people start complaining that they want the old version back, or saying they are "lucky to have saved a backup of the old version of the app", do you?

I guess we'll just have to agree to disagree; I don't see that what I've suggested is horrible. The free version of Pleco exists to sell add-ons; it's a not a gift to the world from a philanthropist. I want Pleco the company to make enough money to remain viable and keep developing Pleco app improvements.

This back-and-forth is quite interesting - thanks - and hints at a number of larger "what do we want to be?" questions that I've been pondering for some time now.

An awful lot of our revenue now comes from a tiny percentage of our active users, those enthusiastic enough to buy more-or-less every dictionary we sell (and, as an added bonus, to tell all of their friends to buy them too), and it's absolutely critical that we not do anything to disappoint that group, or turn away potential future members of that group. So in some respects we are already much more like Temple Run than Angry Birds - instead of getting $1 or $2 from every user, we get nothing from 95% of them and a lot more than $1-$2 each from a small subset. Since those users were the only users we had in the Palm/WM days, it seemed natural to keep them our primary focus on iOS, hence the lack of ads / unobtrusive promotion of add-ons / etc.

But it does feel like we're also leaving a lot of money on the table this way, and even more so in future updates as we hopefully get our UI design to rise above its current "competent but unexciting" and up to the level that gets people to pay for other apps purely on the basis of looks (e.g. weather apps, many of which don't actually do anything more than the built-in iOS weather app but still sell a zillion copies). Assuming we don't want to start charging a few dollars for our basic app, and I don't think we do, and that we don't want to make the prettier design itself a paid add-on (which would probably cost us more in programming time to implement than it would make us in extra sales), it would be nice to find a way to get a little money out of people who are enjoying our app but aren't inclined to pay $10 or $20 for it.

This may simply be a question of making some add-ons cheaper - take some little new features and charge $1 or $2 for them: part-of-speech tags for PLC, say, or a 楷体 font for headwords, or a most-common-few-thousand-characters version of our stroke order module, stuff that still requires users to find their way to the "Add-ons" screen but doesn't require a great leap to purchase once they're there. There's a risk of seeming to nickel-and-dime people, but as long as we give most of it away for free to people who buy / have already bought a bundle, it seems like it would mostly be a positive thing. But I'm not sure how much extra money any of this would actually make.

Basically it boils down to a question of whether people are unaware of our add-ons or just uninterested. If the former, it seems like finding a few small tasteful ways to direct them towards the Add-ons screen probably would help; if the latter, we then have to decide whether to accept that they're not interested and focus on the people who are, or to attempt to get some money out of those uninterested people anyway through ads or cheaper optional features or other such things.

 

[character]

That I'd have to double-check on - Apple actually requires us to write up an official description of each IAP that goes in their database and gets reviewed / approved like any other bit of App Store text, but it's not clear what the requirements are concerning the display of that versus other text that we might come up with instead.

Hopefully you can have something like "Blah Blah Classical Chinese Dictionary" for the approved description and then do A/B testing with other text. Intent is important here; you're not trying to mislead by changing the text, just find out if (say) a bullet list of features is more motivating than a paragraph of text. I guess the alternative would be to have multiple IAP for each item and show different sets of IAP to different users. You could then see how many bought the classical dictionary with description 1 versus 2 versus 3.

An awful lot of our revenue now comes from a tiny percentage of our active users, those enthusiastic enough to buy more-or-less every dictionary we sell (and, as an added bonus, to tell all of their friends to buy them too), and it's absolutely critical that we not do anything to disappoint that group, or turn away potential future members of that group.

While I'm part of that group, there is a downside with it: people who pay once or twice and then use the product for years and years. Pleco is not (currently) like some other products where there are yearly upgrades.

This may simply be a question of making some add-ons cheaper [...] There's a risk of seeming to nickel-and-dime people, but as long as we give most of it away for free to people who buy / have already bought a bundle, it seems like it would mostly be a positive thing.

The examples you give do seem nickle-and-dime-y to me. IIRC, when I was purchasing OCR, the educational discount seemed intentionally hidden. Making that more prominent might limit sticker shock from a lot of your potential customer base.

Basically it boils down to a question of whether people are unaware of our add-ons or just uninterested. If the former, it seems like finding a few small tasteful ways to direct them towards the Add-ons screen probably would help; if the latter, we then have to decide whether to accept that they're not interested and focus on the people who are, or to attempt to get some money out of those uninterested people anyway through ads or cheaper optional features or other such things.

Definitely insert ads in search result listings; otherwise, they will play havoc with your delicately-balanced UI.

I'd not thought of ads in the free version of Pleco, but that's certainly one way to go, if the bulk of your (non-paying) users are in the US. It's my understanding advertising revenue drops off precipitously outside the US. Ads might piss people off who have been using the current free version without ads and feel entitled to such a version.

I know I don't know all the different things Pleco does, so I'm thinking the "what's new in this version" popup would benefit me as well.

While you mull over how to proceed, if the increased download limit allows it, why not go ahead and bundle the manual with Pleco so people will now have detailed descriptions of what all the add-ons do included in the app. Perhaps add a way to see if people ever open the add-ons screen.

 

[mikelove]

Hopefully you can have something like "Blah Blah Classical Chinese Dictionary" for the approved description and then do A/B testing with other text. Intent is important here; you're not trying to mislead by changing the text, just find out if (say) a bullet list of features is more motivating than a paragraph of text. I guess the alternative would be to have multiple IAP for each item and show different sets of IAP to different users. You could then see how many bought the classical dictionary with description 1 versus 2 versus 3.

True, but Apple have never been consistently reasonable about such things. Then again, an awful lot of games seem to use their own UI for this and ignore any Apple-approved text, so that suggests a useful precedent.

While I'm part of that group, there is a downside with it: people who pay once or twice and then use the product for years and years. Pleco is not (currently) like some other products where there are yearly upgrades.

And we're not likely to start doing that, but we would like to get on a more regular schedule of introducing new features and specifically new content to pay for - this is one potential upside of that "Pleco Book Club" idea I floated earlier, subscription content of some sort is a great way to get an ongoing revenue stream from our users. Though we could also just as easily make the subscription for a frequently-and-aggressively-updated dictionary, I suppose.

The examples you give do seem nickle-and-dime-y to me. IIRC, when I was purchasing OCR, the educational discount seemed intentionally hidden. Making that more prominent might limit sticker shock from a lot of your potential customer base.

We did that earlier in the year - there's an "(educ $X.XX)" in every listing now. Though I'm still leaning towards phasing those out.

I'd not thought of ads in the free version of Pleco, but that's certainly one way to go, if the bulk of your (non-paying) users are in the US. It's my understanding advertising revenue drops off precipitously outside the US. Ads might piss people off who have been using the current free version without ads and feel entitled to such a version.

True, which is why we'd want to be offering enough new features for free to hopefully soften the blow. But it's certainly a step I'd rather not take - we're actually doing quite well business-wise at the moment (the slow pace of our iOS update appears to be having little to no impact on sales), and I take enough pride in our app in general that it's probably only something I'd try as a desperation move and not simply as something to fatten our bank balance a bit.

While you mull over how to proceed, if the increased download limit allows it, why not go ahead and bundle the manual with Pleco so people will now have detailed descriptions of what all the add-ons do included in the app. Perhaps add a way to see if people ever open the add-ons screen.

We usually rewrite the manual for a new update while Apple is in the process of reviewing it, so if it had to be ready to go when the update was submitted that would delay our releases by an extra week or so

 

[Alexis]

And a sure-to-be-controversial thought: How about removing the Honesty switch... er I mean the student discount switch, and choosing a mid-range price that reflects the current split of revenue?

This would make the overall prices seem cheaper.

 

[character]

Then again, an awful lot of games seem to use their own UI for this and ignore any Apple-approved text, so that suggests a useful precedent.

I thought Apple doesn't hold itself to following its own precedents in app reviews.

We usually rewrite the manual for a new update while Apple is in the process of reviewing it, so if it had to be ready to go when the update was submitted that would delay our releases by an extra week or so

While you don't want to delay critical bug fix releases, for other releases, I don't think a week's delay is a big deal. If not all the developers are needed for the documentation effort, perhaps they could be working on enhancing the automated test suite during that week, so there's the chance other bugs would be caught before submission.

 

[mikelove]

And a sure-to-be-controversial thought: How about removing the Honesty switch... er I mean the student discount switch, and choosing a mid-range price that reflects the current split of revenue?

This would make the overall prices seem cheaper.

We're considering it - already do that on Android - but the worry is that it may do more harm than good: the majority of our customers now take advantage of that discount, and psychologically, people like to get discounts, so they might actually be more inclined to buy our software because they're paying less than full price for it.

I thought Apple doesn't hold itself to following its own precedents in app reviews.

They've gotten a little better thanks to all the scrutiny - there's an official appeals process now and the appellate circuit does seem capable of applying past precedents absent a policy change that supersedes them.

While you don't want to delay critical bug fix releases, for other releases, I don't think a week's delay is a big deal. If not all the developers are needed for the documentation effort, perhaps they could be working on enhancing the automated test suite during that week, so there's the chance other bugs would be caught before submission.

Not a big deal, but a big enough deal that I don't think the benefits of having an offline manual built-in (as opposed to an online manual that can easily be downloaded for offline viewing) are sufficient to justify it.

 

[character]

Not a big deal, but a big enough deal that I don't think the benefits of having an offline manual built-in (as opposed to an online manual that can easily be downloaded for offline viewing) are sufficient to justify it.

I viewed it as acting as sales material more likely to be looked at than something labeled "Add-ons".

If you don't want to wait for it to be ready, perhaps modify the app so it can pop up something to let users know a new version of the manual is available for download.

 

[mikelove]

I viewed it as acting as sales material more likely to be looked at than something labeled "Add-ons".

Well that's why there's an online manual not in Add-ons - very few people nowadays update Pleco without having an active internet connection on their device. I'm not objecting to the idea of keeping the manual in a prominent location and up-to-date, I'm just not seeing much benefit to having an offline version of it embedded in the app.

If you don't want to wait for it to be ready, perhaps modify the app so it can pop up something to let users know a new version of the manual is available for download.

We're actually planning to do that for all add-ons, since we'd like to tell people about updated dictionaries too.