PodSearch

Under the Radar

Under the Radar 50: Checking the Weather for the Last Time

 

00:00:00   welcome to under the radar a show but

00:00:02   independent iOS app development I'm

00:00:04   Marco Arment and I'm David Smith under

00:00:06   the radar is never longer than 30

00:00:08   minutes so let's get started so this

00:00:11   week I had a sad thing happened to me

00:00:14   that I wanted to unpack on the show so I

00:00:16   had to retire one of my apps and one of

00:00:20   the apps that obviously I have a lot of

00:00:21   apps and I've quietly retired many of

00:00:24   them in the past the ones that never had

00:00:26   any downloads the ones that just didn't

00:00:28   work out and you kind of just removed

00:00:30   them from the store and that's that but

00:00:32   I had one of them had removing out from

00:00:34   the store that wasn't one of those apps

00:00:36   that the had some success that was

00:00:38   something I was proud of being out there

00:00:40   I think this is actually the first app

00:00:42   of yours that I've ever both heard of

00:00:44   and downloaded that you're also now

00:00:46   killing the hakko it comes full circle

00:00:49   and the app was it's called check the

00:00:51   weather and it is a weather app that at

00:00:56   the time I launched it and I know

00:00:58   exactly when I was it was for week four

00:01:00   years ago as of last Monday the reason I

00:01:03   and I and sort of publicly announced its

00:01:05   end of life last Monday for that reason

00:01:07   when I was planning this and scheduling

00:01:09   it out and I looked at the schedule and

00:01:10   I was like wait I think I launched it a

00:01:12   wrestler or somewhere in middle of

00:01:14   October and it turned out it was for you

00:01:17   exactly four years ago so the app had a

00:01:19   four year life oh and it's just a you

00:01:21   know it's a weather app it's a strange

00:01:24   thing in some ways to try and describe

00:01:26   what it is or what it does what made it

00:01:27   special because the weather I have is

00:01:29   like they're all strange in the sense

00:01:31   that they all have the same basic data

00:01:34   and then the differences between all the

00:01:36   apps are just you know questions of

00:01:38   style of nuance but check the weather

00:01:43   always has kind of a special place for

00:01:44   me because it was the first app that I

00:01:47   ever felt that I kind of did it right no

00:01:50   you know it's it I started from the

00:01:52   beginning and said you know what what

00:01:54   happens if I take a run at a very

00:01:56   well-established category doing

00:01:59   everything right you know so doing

00:02:00   having a proper actual marketing plan

00:02:02   that was an actual thing that I did

00:02:04   before either than just kind of throwing

00:02:05   it over the wall or sending some blind

00:02:07   emails

00:02:08   and it was localized from day one it had

00:02:12   great voiceover supported all the things

00:02:14   that I kind of was about in my mind were

00:02:16   always the things that you had to do to

00:02:19   launch an app that was you know if

00:02:21   you're gonna do it right if you're going

00:02:22   to give it an honest chance and if

00:02:25   you're curious if your listener to this

00:02:26   show but you weren't a listener to the

00:02:28   podcast I used to do called developing

00:02:30   perspective I'll have links in the show

00:02:32   notes to I did a whole series of

00:02:34   episodes on developing perspective which

00:02:37   rather than being 30 minutes long is

00:02:38   only 15 minutes long so even though this

00:02:40   is a lot of episodes it wouldn't take

00:02:42   that long to catch up we're actually

00:02:43   walked through at the in real time as I

00:02:45   was building back in 2002

00:02:48   back in 2012 I walked through the

00:02:51   process of what went into this but

00:02:52   anyway the long and the short of it is I

00:02:54   had this app I launched it it did great

00:02:57   it had a very big spiky launch I think

00:02:59   it peaked at number three overall in the

00:03:02   App Store which at the time for me was

00:03:05   just you know mind-bending I actually

00:03:06   have very warm and specific memories of

00:03:09   that day when it launched just

00:03:11   celebrating having my wife found someone

00:03:14   to watch the kids and came into the

00:03:15   office with I think even Chuck's like

00:03:19   Chinese for dinner and a bottle of

00:03:20   champagne and we you know actually

00:03:22   celebrated exceda it all kind of worked

00:03:24   and it kind of all worked out but that

00:03:27   was four years ago

00:03:28   that's a long time ago in App Store

00:03:31   terms and so this week I now it will no

00:03:35   longer be supported and for something

00:03:37   like a weather app that means I'm

00:03:39   turning off the API at the backend and

00:03:43   that's kind of a sad thing but it was

00:03:46   kind of inevitable because the tricky

00:03:48   thing with someone Apple I checked the

00:03:49   weather is that at least the way that I

00:03:51   launched it is at the time we didn't

00:03:53   really have a lot of this kind of

00:03:53   subscription options you have we have

00:03:55   now and so I was really more sort of

00:03:57   like freer page where my options four

00:03:59   years ago I could have sort of done

00:04:01   subscriptions with some kind of you buy

00:04:03   an in-app purchase and then you buy

00:04:05   another one and then you buy another one

00:04:06   but there was certainly wouldn't have

00:04:07   been able to be Auto renewing so I made

00:04:09   it a free app and or so I made it a paid

00:04:11   app and I put that out in the air so I

00:04:14   had a whole lot of people who gave me

00:04:15   money

00:04:16   four years ago I think about a third of

00:04:18   the users bought it the first day you

00:04:21   know had a nice big

00:04:22   like he launched but most of you they've

00:04:24   never subsequently got it given me any

00:04:27   revenue and ongoing costs are really

00:04:31   rough especially when they aren't just

00:04:34   you know two or three Linode boxes

00:04:36   things like weather data and even fonts

00:04:39   actually like I had some custom fonts in

00:04:41   this application that had annual

00:04:43   subscriptions that I have to pay that's

00:04:45   the worst tell you one thing I mean like

00:04:48   you know server costs like you kind of

00:04:50   can't get around that but as somebody

00:04:52   who has licensed a lot of fonts for apps

00:04:55   over the years anything you can do to

00:04:57   avoid ongoing recurring costs in your

00:05:00   app you should do that and if you have a

00:05:02   fancy font you know try it try to get

00:05:04   one that you can license that for a flat

00:05:06   fee especially you know these days you

00:05:09   can often get decent fonts for either

00:05:11   free or for maybe a few hundred dollars

00:05:13   up front flat forever and that's if you

00:05:16   can do that that's great

00:05:17   sure because it's any of these things

00:05:20   like it was always a strange thing

00:05:22   because it forced me to every year I had

00:05:24   to sit down and decide do I still want

00:05:26   to pay for this font to this point the

00:05:28   opposite kind of a I followed the usual

00:05:30   trajectory like it had that big spike at

00:05:32   the beginning and then all you know kind

00:05:33   of felt almost nothing and it became

00:05:36   kind of like a charity project because

00:05:38   you know like I still used the app and I

00:05:40   mean through this week I was still using

00:05:42   the app but at some point I had to

00:05:44   decide you know did every year when I

00:05:46   get my email from font foundry and says

00:05:48   hey you know would like to charge you

00:05:50   another three or four hundred dollars

00:05:51   it's like great and every month when I

00:05:54   get my bills from the radar provider and

00:05:57   the weather data provider it's like mm

00:05:59   hmm you know this is fine I mean the

00:06:02   costs involved weren't killer at least

00:06:04   up until recently and that's part of why

00:06:06   I have to shut it down now because the

00:06:09   there's a change at pricing changed with

00:06:11   my radar provider that means that it

00:06:12   would just have gone from like oh it's

00:06:14   kind of fun it's a little charity

00:06:15   project to something that would have

00:06:18   just been kind of absurd to keep going

00:06:20   well I think I think ultimately the you

00:06:23   need to think about like you know what

00:06:25   and I think we'll get to this later like

00:06:26   what do you owe your users because in my

00:06:28   opinion the second you are losing money

00:06:31   like there is no acceptable amount of

00:06:33   money to be losing on and on

00:06:35   gong basis four and a half like to me

00:06:38   once you're losing money that's it it's

00:06:40   it's gone sure and I definitely don't

00:06:42   see you can see where you're coming from

00:06:43   like it is such a hard thing though from

00:06:46   the emotional perspective like what

00:06:48   you're saying makes total sense that you

00:06:49   know like if I'm running a business and

00:06:50   I have a division of my business that is

00:06:54   continue is making a loss with no

00:06:56   prospect of turning that around the

00:06:59   logical like good business move is to

00:07:02   just shut it down and move on but it's

00:07:06   hard when you kind of have something

00:07:07   that you feel more emotionally invested

00:07:09   in where it feels like oh no you know

00:07:12   this is like with my losses for this app

00:07:14   was my first big break like I want to

00:07:15   you know I feel sad for it to go and so

00:07:19   I think I definitely in this case I

00:07:21   probably did let it go longer than in

00:07:23   retrospect I should have and not

00:07:25   necessarily even just for the money

00:07:26   reasons so I think what you're saying

00:07:27   makes sense but even just in retrospect

00:07:29   as I think of the like the the burden it

00:07:33   places on you cognitively when you have

00:07:35   something that you're sort of you have

00:07:37   this affinity for and you're kind of

00:07:39   proud of but it isn't really going well

00:07:42   and you know every time an iOS update

00:07:44   comes out it's like mm should I do this

00:07:46   to check the weather

00:07:47   you know like Apple like the Apple watch

00:07:49   came out and I was like oh there's some

00:07:50   really cool things you could do for you

00:07:51   know weather app on the Apple watch

00:07:53   I was like man the app just doesn't make

00:07:55   money like it just really isn't getting

00:07:57   traction and I don't know if just

00:08:00   putting more effort into it

00:08:01   is we really going to get the outcome

00:08:02   but then everyone every every time

00:08:04   there's a big thing and have that that

00:08:06   sense of doubt I feel like that's the

00:08:07   kind of thing that honestly more than

00:08:09   the money is a bit better an incentive

00:08:11   to make sure that you know if when

00:08:13   something is clearly not going anywhere

00:08:15   that you just kind of say yep it's done

00:08:17   and move on anyway so that is why I have

00:08:21   gotten to the point that I had to pull

00:08:22   check the weather from the store and

00:08:25   it's probably worth mentioning a little

00:08:28   bit about kind of what that process

00:08:29   looks like

00:08:30   mechanically before we get into some of

00:08:33   the kind of the implications and things

00:08:35   that come out of that so when you decide

00:08:38   that you want to pull an app from the

00:08:39   App Store it's surprisingly easy to do

00:08:41   you just go into iTunes Connect you

00:08:44   click on the app there's no think it's

00:08:45   availability and pricing is the area in

00:08:47   the App Store that you can go into

00:08:49   all right in iTunes Connect that you go

00:08:50   to and say you know remove says you

00:08:51   removed from sale you used to have to go

00:08:53   in there and specifically select every

00:08:57   country that you didn't want it in until

00:08:58   use pet so let me select at the mall now

00:09:00   you can just push a button I think that

00:09:02   says remove some move from sale and that

00:09:05   worked pretty well it was gone from the

00:09:07   App Store and then you have the a more

00:09:10   interesting question perhaps of how you

00:09:12   communicate to your users so as most

00:09:15   people who download your app don't know

00:09:17   who you are they don't follow you on

00:09:19   twitter they don't read your blog and I

00:09:22   kind of struggled with this in this case

00:09:24   like I've certainly done the I have a

00:09:25   blog post mentioned on Twitter to kind

00:09:27   of put it out as best I could for most

00:09:30   of your users either you need to do a an

00:09:34   app update and just mention this in the

00:09:36   release notes that the app is going away

00:09:39   or you would probably need to have some

00:09:42   mechanism built into the application to

00:09:45   reach out to them some kind of messaging

00:09:46   scheme at this point the app has so few

00:09:49   users that I didn't decide that

00:09:50   something like that was worth doing of

00:09:53   like building a special build of the app

00:09:54   and putting it out there that popped up

00:09:56   a window that said hey your app your

00:09:58   data is gonna go away that's something

00:10:00   that you'd kind of would have to do if

00:10:01   you were you know if we're ever ayat e

00:10:04   of reasons there a variety of reasons

00:10:05   that you would have to turn something

00:10:06   off and in this case it's turning it off

00:10:09   because the user ship is fallen off

00:10:11   dramatically but I've certainly have

00:10:13   known have friends who ended having to

00:10:15   end a product in some ways because it

00:10:18   was too successful that it kind of had

00:10:20   gotten ahead of itself in terms of

00:10:21   usership

00:10:22   but didn't have a revenue model to back

00:10:24   it up and so a certain point they're

00:10:25   just like well I just have to turn it

00:10:27   off because the money is just kind of

00:10:29   running away from me so in that case you

00:10:30   probably have to build something into

00:10:31   the app to turn it off and then you need

00:10:35   some kind of mechanism to actually like

00:10:37   physically turn it off in this case

00:10:39   because I'm not hosting much myself

00:10:42   there's not like I have servers an

00:10:43   infrastructure that I have to physically

00:10:45   unwind I'm mostly I'm just stopping

00:10:48   paying for things and then the ultimate

00:10:51   providers will turn it off but it is

00:10:53   definitely something that I would say

00:10:55   that I never really thought of when I'm

00:10:57   you never cease e to not think about

00:10:58   when you're making something but that

00:11:01   what I would I would

00:11:02   she'd do if I needed to turn this off

00:11:04   you know like at some point if some of

00:11:06   my other projects that have Web Services

00:11:07   that I run were to go away it's kind of

00:11:10   a tricky question to be like well what

00:11:12   do I do with the data for example like

00:11:14   how do I safely and securely turn

00:11:16   everything off it's the kind of thing

00:11:19   where it can so easily get away from you

00:11:20   when you're developing like you can just

00:11:22   add things like oh this will be fine

00:11:23   this will be fine I just not even think

00:11:24   about like all the different services

00:11:25   you're starting up you know API

00:11:27   memberships you're sorting all the

00:11:28   different providers and everything and

00:11:30   it can get away from you pretty fast oh

00:11:31   sure I mean it's probably also worth

00:11:33   saying that a lot of these same

00:11:34   considerations come into play if you

00:11:36   ever transfer an app oh yeah so if you

00:11:38   ever sell an app it's something that's

00:11:40   probably worth it's just sort of as a as

00:11:42   a pro tip is keeping track in a document

00:11:44   or somewhere it just at a high level all

00:11:47   of the various things that you've ever

00:11:50   paid for or bought associated with the

00:11:53   development of an app in terms of yeah

00:11:55   like keeping track of like there's a

00:11:57   domain associated with this app you know

00:11:59   if you go to check the weather dock oh I

00:12:01   think it was like there's a domain and I

00:12:03   would need to remember to go and turn

00:12:05   off auto renew on that domain there's

00:12:08   all you know that were I think there's

00:12:10   three or four different weather data

00:12:12   providers that I have a created accounts

00:12:15   with at some point for this app and

00:12:16   while as you know if in some ways the

00:12:19   last some of them would just kind of

00:12:21   turn off on their own if they're

00:12:22   usership went to zero the reality is I

00:12:25   also should probably go and actually you

00:12:28   know terminate my account there so I'm

00:12:30   not don't have all of these open these

00:12:32   little open things in the future that

00:12:33   could come back and bite me in some

00:12:36   weird way and on the flip side on the

00:12:38   positives say hey the Apple had been

00:12:39   acquired and been shipped over to

00:12:40   somebody you're gonna have to be able to

00:12:42   give the new owner at the list of like

00:12:45   here's all the things that I actually

00:12:46   have to give you which is way more than

00:12:49   just the source code to the app where's

00:12:52   Pasha this week by Pingdom go to Pingdom

00:12:54   comm slash radar to start monitoring

00:12:56   your websites and servers today Kingdom

00:12:59   offers a 14-day free trial and when you

00:13:00   enter offer code radar at checkout you

00:13:02   get 20% off your first invoice Pingdom

00:13:05   is focused on making the web faster and

00:13:06   more reliable for everyone who has a

00:13:08   site or service they offer powerful

00:13:10   easy-to-use monitoring tools and

00:13:13   services so if you're opinio music you

00:13:14   can monitor the availability and perform

00:13:16   your servers databases websites all very

00:13:19   very easily they take care of this by

00:13:21   using more than 70 global test servers

00:13:23   that emulate visits to your site they

00:13:25   can check its availability often as

00:13:26   every minute and they can a different

00:13:28   test with that things like they can use

00:13:29   a cookie to log in they can follow forms

00:13:31   they can check for certain strings on

00:13:34   the resulting page they can check for

00:13:36   certain timeouts Pingdom is incredibly

00:13:38   great I use it to monitor all my stuff

00:13:40   and I've used it since the tumbler day

00:13:42   is very very long time it goes since

00:13:43   long before they were response here

00:13:44   before I even started podcasting at all

00:13:46   I've been using them for a very very

00:13:48   long time and I have had nothing but

00:13:50   great experiences with Pingdom that's

00:13:53   why I can personally recommend them even

00:13:54   they cannot pay me to say that I can

00:13:55   actually say that directly they are

00:13:57   awesome and you should be using them so

00:13:59   payment makes it very very easy to

00:14:00   monitor your sites all you need to do is

00:14:03   give them a URL that you want to monitor

00:14:04   and the various optional conditions you

00:14:06   want to check for and they take care of

00:14:08   the rest when they detect an outage you

00:14:10   can be immediately alerted so you can

00:14:12   fix the error before the downtime

00:14:14   affects you you don't want to be caught

00:14:16   with you know your site down and you

00:14:18   don't even knowing it and people on

00:14:20   Twitter telling you hey by the way you

00:14:21   know your site's been down for the last

00:14:22   couple hours no you want to be the first

00:14:24   person to learn and with a Pingdom you

00:14:26   will be the first person notified so you

00:14:27   can fix it before anybody else really

00:14:29   sees it go to Pingdom comm slash radar

00:14:32   for a 14-day free trial and use offer

00:14:34   code radar to get 20% off at checkout

00:14:37   thank you very much - Pingdom response

00:14:38   during our show so the other kind of

00:14:40   consideration or thing that i Eve this

00:14:42   is all this this process it made me

00:14:43   think of because it's probably only sort

00:14:45   of helpful to talk about the actual like

00:14:47   this the instances of this app but if I

00:14:49   take a step back and think about just in

00:14:52   general as I'm developing an app are

00:14:54   there lessons that I can learn from this

00:14:56   that I think would be useful going

00:14:59   forward and the first one that I was

00:15:01   trying to think of and came up with was

00:15:02   the importance of being careful about

00:15:05   what I'm communicating or promising to

00:15:09   my customers as I start out as I feel

00:15:14   like it's easy when you're starting out

00:15:15   and everything's kind of rosy and nice

00:15:18   to get a little carried away and kind of

00:15:22   make commitments that are down the road

00:15:24   going to be more difficult for you to

00:15:26   follow through on and you know it's a

00:15:29   tricky thing maybe on this case

00:15:30   the pricing where it's really not a

00:15:32   sustainable business model for an app

00:15:33   that has an ongoing cost to have a

00:15:35   one-time purchase upfront but in the

00:15:38   early days of the app I kind of gave the

00:15:40   impression whether implied or actual

00:15:42   enough to actually go back and look at

00:15:43   my communication to see what I actually

00:15:45   said but I kevin lee implied that you

00:15:47   know you buy this app and you have

00:15:49   weather data and there was no end date

00:15:52   put on that and that creates some really

00:15:55   awkward things that kind of kind of feel

00:15:57   like they tie my hands whether

00:15:59   necessarily or just because of the way

00:16:01   that i feel going forward that kind of

00:16:04   you know reduce my options because in

00:16:06   this case you know i said about a third

00:16:07   of the users bought it in one day so

00:16:10   whatever i communicated to those people

00:16:12   that first day when everything was

00:16:13   cheerful and rosy and i was so

00:16:15   optimistic about its future I kind of

00:16:19   have to live with going forward and it

00:16:21   definitely kind of makes me think of as

00:16:23   I launch things now or as I move forward

00:16:26   well there's also good for me to be

00:16:28   thinking about you know this is

00:16:29   sustainable and so on but it's even the

00:16:31   importance of being careful about what I

00:16:33   say that I'm not making pointers or

00:16:35   promises into the future it's kind of

00:16:38   like any time I see an accompanying that

00:16:40   now says something's unlimited or even

00:16:44   lifetime subscription or those types of

00:16:47   words I am immediately skeptical I

00:16:50   immediately think no that's you that

00:16:52   can't be right

00:16:54   you know you can't use those words with

00:16:56   actual honesty because you can't give

00:16:59   someone an unlimited something or a

00:17:01   lifetime of something because you're

00:17:03   making promises into the future that you

00:17:05   really can't stand behind and that's

00:17:08   probably a dangerous thing for me you

00:17:10   know for especially a small developer to

00:17:11   ever do well especially goes like when

00:17:13   you're when you have a paid upfront app

00:17:14   or you know any kind of and Wow however

00:17:16   people are paying for it if somebody

00:17:17   pays for your app there many people

00:17:20   maybe I don't even meet maybe even most

00:17:22   people have this expectation at the

00:17:24   moment of payment that I am paying for

00:17:26   this app for it to work forever and of

00:17:28   course in reality that's you know that's

00:17:30   not how anything in this industry works

00:17:32   but that is kind of like the mindset and

00:17:34   that's why people get so upset when an

00:17:37   app that they paid for two years ago all

00:17:39   of a sudden gets discontinued or

00:17:42   requires a paid upgrade to keep working

00:17:44   on

00:17:44   the newest whatever's you know people

00:17:46   they when they pay upfront as something

00:17:49   that's presented as a one-time fee they

00:17:51   really do assume even if it's

00:17:54   subconscious

00:17:55   they see that purchase as a lifetime

00:17:58   purchase when in fact that is impossible

00:18:00   to sustain I mean you know people

00:18:02   basically software needs ongoing revenue

00:18:06   streams because people expect ongoing

00:18:10   updates ongoing service ongoing

00:18:13   compatibility fixes and even ongoing

00:18:15   improvements and yet they pay once

00:18:18   upfront and and they and they expect all

00:18:20   those things to come for free forever so

00:18:22   like some kind of recurring subscription

00:18:24   or recurring revenue model but you know

00:18:26   whether its description or ads or other

00:18:27   other things the reason why this is so

00:18:30   helpful is because like that matches up

00:18:31   more to people's expectations of what

00:18:33   the product will do for them and how the

00:18:34   product will keep moving but you have

00:18:36   something like you had to check the

00:18:37   weather because it was about like three

00:18:38   or four bucks up front I think it was

00:18:40   yeah yeah so it was like you know

00:18:42   whatever it is three or four bucks up

00:18:42   front then you know people who pay three

00:18:45   or four bucks two three years ago you

00:18:47   know they're gonna expect that app to

00:18:49   still be working now if they're still

00:18:50   using it and you're going to get some

00:18:52   people who are very upset about this

00:18:54   discontinuation but I don't really see

00:18:56   how you could have kept it going

00:18:58   otherwise I mean if you had I mean we

00:19:01   could talk about different business

00:19:01   models and stuff but you know if it was

00:19:03   subscription funded from the very

00:19:05   beginning that would be another thing if

00:19:06   it were ad funded then you know it would

00:19:09   it would just kind of keep making money

00:19:11   through usage hopefully enough to cover

00:19:13   it but that's not even a guarantee and

00:19:16   you know so if there was ongoing revenue

00:19:18   you could you could possibly sustain it

00:19:20   more but people really do you know when

00:19:23   it comes to you know you know geeks like

00:19:25   us know to to doubt claims of unlimited

00:19:28   or a lifetime but when you have paid

00:19:31   upfront that is literally what people

00:19:32   expect they expect it to be a lifetime

00:19:34   purchase even though they might not care

00:19:36   in a year or two but the ones who still

00:19:39   use it they will care sure and I think

00:19:42   beyond even just the consideration of

00:19:44   like what's the best business model and

00:19:46   obviously like in retrospect I think of

00:19:48   myself four years ago being very naive

00:19:50   that this kind of would have worked

00:19:53   and I think the important thing there is

00:19:55   - it's a to think it through but like I

00:19:58   got so caught up in the actual process

00:20:00   of building it and thinking that it was

00:20:01   cool I don't think I thought four years

00:20:04   into the future about what would happen

00:20:06   and then me even if you don't have a

00:20:09   good alternative for the business model

00:20:11   like even if I'd still launched the app

00:20:12   you know paid up front app I likely

00:20:14   should have given some thought to how I

00:20:19   talked about it and how kind of the

00:20:21   expectations that I said yeah maybe I

00:20:22   should have said you know the app

00:20:24   includes weather data for two years and

00:20:26   maybe I don't have a good answer for

00:20:27   what happens in two years but they're

00:20:31   sort of the expectation that maybe down

00:20:33   the road there will be some more money

00:20:34   that you'll have to pay pay for me to

00:20:36   keep doing it and communicating that

00:20:38   rather than kind of if you don't say

00:20:40   anything like you're saying the

00:20:43   customers expectation is that it's like

00:20:45   well I make one payment and then it's

00:20:46   the rest of my life for as long as I

00:20:48   want to use it you know I will be

00:20:49   sitting there ready for them to help

00:20:51   them as however that goes well and for a

00:20:54   while that actually works for a lot of

00:20:57   apps like especially in the earlier days

00:20:58   of the apps were like four years ago and

00:21:00   you long says that was a totally

00:21:01   reasonable thing to do like when I

00:21:02   launched instapaper and you know day one

00:21:04   be Happ store in 2008 well it was more

00:21:07   like day 3 whatever doesn't hate it was

00:21:11   just an upfront paid app it was first 10

00:21:14   bucks then about a year later lowered it

00:21:16   to 5 bucks and stay that way for a long

00:21:18   time and the idea you know I was running

00:21:20   this big expensive web service behind it

00:21:22   but every month I just kept selling more

00:21:24   copies of the app and so the growth was

00:21:27   just funding it like just the additional

00:21:29   new copies that were selling every month

00:21:31   wasn't you know that was enough to fund

00:21:33   the app and overcast first year I did

00:21:35   the same thing with you know with with

00:21:37   the five dollar one-time on and I

00:21:38   purchased to unlock the the you know the

00:21:40   good features the same thing and it

00:21:42   worked fine for a little while but as

00:21:45   the App Store has gone on that that

00:21:47   incredible growth that we had in the

00:21:48   earlier years is harder and harder to

00:21:51   come by and the money you're getting

00:21:53   from each person is like is often going

00:21:55   down as we've had to you know lower

00:21:57   prices to be more competitive in the

00:21:58   market and so now you you really have to

00:22:02   think from the beginning how am I going

00:22:04   to get ongoing revenue from this whereas

00:22:06   even

00:22:06   as recent as four years ago when you

00:22:09   launched this we really didn't have to

00:22:10   think about that it wasn't you being

00:22:11   naive it was actually the conditions of

00:22:13   the market at the time you could just

00:22:16   fund an app entirely by its own growth

00:22:18   sure and you know it's how the good old

00:22:22   days but I mean never nevertheless like

00:22:24   the reality is it's something that I

00:22:26   mentioned here as a cautionary tale both

00:22:29   for myself and others that if you don't

00:22:32   think about this upfront eventually

00:22:35   you're going to have to deal with it I

00:22:37   mean it reminds me in a weird way of

00:22:38   like contract negotiation where I

00:22:40   remember getting the best advice I've

00:22:42   sung to a guy who was well he was my

00:22:43   lawyer but in this case he wasn't acting

00:22:45   as my lawyer because he's you know a

00:22:46   lawyer he was telling me like the reason

00:22:48   he likes doing contracts for developers

00:22:52   is his job he felt was to think through

00:22:56   all of the worst-case scenarios for them

00:22:59   that they don't want to think about that

00:23:01   they're sitting down with the client and

00:23:02   they're super excited like it's the

00:23:05   first big client they've ever gotten or

00:23:07   whatever and they're really excited

00:23:08   about it and they can talk about all the

00:23:10   fun things they're gonna do and the

00:23:11   awesome work they're gonna you know be

00:23:12   able to do and collaborate together and

00:23:15   you just don't think about these things

00:23:17   that what do we do do you know in in the

00:23:20   sense of a contract like what happens to

00:23:22   the code at the end of this project what

00:23:23   happens if you're sick and unable to

00:23:25   continue working what happens if you

00:23:28   just have massive disagreements and

00:23:30   don't get along like those kinds of

00:23:32   contingencies those kinds of things that

00:23:34   are you hope you never really hope for

00:23:36   you just hope it is always going to be

00:23:38   the best case you know in this case with

00:23:39   check the mother if the sales had just

00:23:40   kept going forever it would the Apple

00:23:43   just keep going forever but if you force

00:23:45   yourself to think about the weird

00:23:47   not-so-great things up front you can

00:23:51   just plan for them or at least have some

00:23:52   sense of what you might do and it

00:23:55   probably would have helped in this case

00:23:56   would have helped me avoid kind of it

00:23:58   feels a bit awkward and it feels a bit

00:24:00   more like I'm letting people down and

00:24:02   generally people have been really

00:24:03   gracious about it I think the nice thing

00:24:04   about a third of the users buying it in

00:24:06   the first day and probably about half of

00:24:09   the users buying it in the first month

00:24:11   is that that was all four years ago and

00:24:13   so I think they've stopped using the app

00:24:15   or if they have used it they've been

00:24:16   using it for four years and paying you

00:24:18   know less than a dollar a year is

00:24:20   some reasonable to them but if anything

00:24:23   it's just the cautionary tale of think

00:24:25   about these things think about what's

00:24:27   what this would look like down the road

00:24:28   and have if not a plan in specifics at

00:24:31   least a plan in general for how you're

00:24:33   gonna deal with it the other thing that

00:24:36   I've kind of been running into as I've

00:24:38   been unraveling it that it's kind of a

00:24:39   lesson that that would be worth talking

00:24:42   about is as much as I feel like I think

00:24:45   of the AppStore and the way people use

00:24:47   my app as only ever using the latest app

00:24:50   on the latest version of the OS which is

00:24:53   in general like surprisingly true as

00:24:56   I've been kind of unwinding this app and

00:24:59   working out what I need to turn off I

00:25:00   always went back and looked at old users

00:25:04   of the application and this is where

00:25:05   things get funny and why like I've

00:25:07   changed data providers for example and

00:25:09   I'm not sure and I need to make sure

00:25:12   that I go and turn off those first

00:25:13   accounts that I used to get weather data

00:25:15   from because people are still using the

00:25:17   app that hasn't you know really really

00:25:20   really old versions like I went and

00:25:22   looked in my analytics and I have people

00:25:23   who are using the app that was published

00:25:25   in 2013 still like as of a couple of

00:25:30   days ago I was you know still having

00:25:32   people who are using like hitting my

00:25:33   server hitting my web services with that

00:25:36   you know with that version string is

00:25:37   this like from people who using really

00:25:39   old iOS versions yes

00:25:41   it's usually it's if that's that is

00:25:43   usually the situation that it's like

00:25:44   somebody has an old iPod Touch or a an

00:25:49   old iPhone that can't be upgraded beyond

00:25:51   a certain point and I think like the

00:25:53   version they were using there was like

00:25:55   it's the last version that worked with

00:25:57   iOS 6 or with iOS 5 even like I mean

00:25:59   it's getting really far back in time and

00:26:03   so they're just stuck using that version

00:26:04   and if they're still using it

00:26:06   like in some ways you know good for them

00:26:07   that's that's fine but in yeah I usually

00:26:11   don't worry about those people in terms

00:26:13   of new development but when you turn

00:26:14   things off it is definitely something

00:26:17   that I felt like I have to think through

00:26:19   now is are there any things that old old

00:26:23   versions of the app do that I need to

00:26:26   think about now and in a weird way it

00:26:30   gives me thought to being careful

00:26:33   being thoughtful as I develop about what

00:26:35   I wear I put things in the app and

00:26:37   reminds me there's like physical feature

00:26:38   there's a mechanism in check the weather

00:26:40   to dynamically change some of the API

00:26:45   keys inside of the application so that

00:26:48   if I you know had a problem with where

00:26:50   the weather provider I could some in

00:26:52   some ways kind of switch things around

00:26:54   dynamically and it does that you know

00:26:56   through a web service and I'm glad it

00:26:59   I've definitely used that a few times

00:27:00   and especially for this is a good

00:27:02   reminder for me of building mechanisms

00:27:05   like that into our apps is probably a

00:27:06   good idea to at least to some degree

00:27:08   because you can't always rely on the App

00:27:12   Store updates being the way you change

00:27:14   that you know if something breaks I have

00:27:15   to go and submit it and even though app

00:27:17   review is amazing like recently I had an

00:27:19   up I had an update yesterday where I

00:27:21   submitted it in the morning in about

00:27:23   three hours later it was approved which

00:27:24   is crazy and awesome but that's only

00:27:27   ever gonna capture like 90 percent of

00:27:29   users and so at least give some thought

00:27:31   to what happens that ten percent of

00:27:33   users yeah it's also really nice to have

00:27:35   from day one a mechanism server-side to

00:27:38   to show a a message like show an alert

00:27:41   box to everybody with with server-side

00:27:43   provide a text I'm and you could have it

00:27:45   be like you know every time or just a

00:27:47   one time showing but somehow have

00:27:49   something in the in there in your apps

00:27:51   so that you could you know you know at

00:27:53   this point rather than have it you know

00:27:54   once everything shuts down rather than

00:27:56   the app just launching and erroring out

00:27:58   with some kind of server error that it

00:28:00   like you know and no internet connection

00:28:01   anything that's like that's not really

00:28:02   right you know then to have to have a

00:28:05   different method where you couldn't you

00:28:07   can like put a file and s3 somewhere

00:28:08   that basically that the app checks and

00:28:10   says are and then you can show a box

00:28:11   that says like sorry we've had to shut

00:28:13   down this app because of what blah blah

00:28:14   blah you know it's something that I have

00:28:17   occasionally built into my apps and I've

00:28:19   been very thankful every time I have I

00:28:21   think the last thing is the thought I

00:28:22   have is the as I finish up this app it's

00:28:26   a weird thing because I still have ideas

00:28:30   for what a cool weather app could be and

00:28:32   do and I had to go navigate a little bit

00:28:36   the thought of if I made another one

00:28:37   would it make people mad or feel like

00:28:39   they've I've cheated them out of

00:28:41   something or I should have just put that

00:28:42   into the app that they've already paid

00:28:43   for

00:28:44   but obviously the reality is and this is

00:28:46   the sort of the end of the

00:28:47   that I came to is you know if I decide

00:28:51   to make a weather app again it'll just

00:28:54   be a new thing and hopefully I may if

00:28:55   the reality is I just need to take the

00:28:57   lessons I learned from this put it into

00:28:59   it you know and that will be okay and

00:29:02   it's being careful of not making choices

00:29:06   just to avoid people being mad at you

00:29:08   because people are always gonna be mad

00:29:10   at you for something yeah and so in this

00:29:13   case I was having thought like you know

00:29:14   I don't know why I was part of why I

00:29:16   didn't pull check the weather is that at

00:29:18   some point I've always thought maybe a

00:29:19   you know I have these cool ideas for

00:29:21   weather app maybe I'll put them in to

00:29:22   check the weather and then I sort of

00:29:24   keep it around just to avoid people

00:29:26   being mad the reality is if I have those

00:29:28   ideas and I did some point down the road

00:29:30   I want to do it I should just do it and

00:29:32   if people are mad they're mad but I

00:29:34   won't be I'll be happy and I'll be glad

00:29:35   and I'll be making something that I'm

00:29:37   excited about and so that's just still

00:29:40   an encouraging thought that I get gave

00:29:42   myself this thing of this week is I went

00:29:44   in and had to hit the button to say you

00:29:46   know it's it's it's finally time to put

00:29:48   this to bed next time use a free font

00:29:50   there you go alright thanks for

00:29:52   listening everybody and we'll talk to

00:29:53   you next week

00:29:54   bye