0:00:00welcome to under the radar a show about
0:00:02independent iOS app development I'm
0:00:04Marco Arment and I'm David Smith under
0:00:07the radar is never longer than 30
0:00:08minutes so let's get started
0:00:09so last week Apple launched App Store
0:00:12search ads finally after we first heard
0:00:14about them in June now they are launched
0:00:16they're running and you can buy them and
0:00:18we have and we like to share our
0:00:20experience with that and I think we've
0:00:22had I've had a lot of questions from
0:00:23developers about just what some of the
0:00:25terms mean with what you're buying
0:00:27they're reporting that you get from the
0:00:29App Store search ad system so I thought
0:00:31first we go over a quick introduction to
0:00:33the basic terminology of buying and
0:00:36selling ads on the Internet
0:00:38and this will lead into specifics about
0:00:40the App Store search ads so first you
0:00:43know I wanted to first go over some very
0:00:45basic stuff so what is you know things
0:00:47like impressions taps conversions so
0:00:50very very basically here I'll try to go
0:00:51through this relatively quickly an
0:00:53impression is just how often an ad is
0:00:54shown it's one showing of an ad to one
0:00:57person click or tap that's obvious and
0:01:00then a conversion is when somebody who
0:01:03has clicked on or tapped on that ad then
0:01:06does what the advertiser wanted so
0:01:08usually it's buying something in a store
0:01:09or it's like creating signing up for a
0:01:11new account on a site or a service or in
0:01:14the case of search ads for apps
0:01:15installing the app so that is the
0:01:17conversion is you know and ads are
0:01:19priced in different ways one of the most
0:01:22common kind of the baseline is CPM which
0:01:24stands for cost per Mille a which is the
0:01:27French word for a thousand which I'm
0:01:28probably mispronouncing it basically
0:01:30cost per thousand impressions so caught
0:01:33how much you pay per thousand people who
0:01:35see the ad the CPM model strongly favors
0:01:39publishers generally in like where it
0:01:42places the risk if I'm showing the ad
0:01:44and somebody sees it but they decide not
0:01:46to click on it or they decide not to go
0:01:48through and buy it who bears that cost
0:01:50of somebody seeing the app but just the
0:01:52ad not working with CPM the publisher
0:01:55gets paid regardless of whether the ad
0:01:56has has you know worked it does however
0:01:59have some weird incentives it
0:02:01incentivize the publishers to put as
0:02:02many ads on screen at spot as possible
0:02:05that's why you have pages that are
0:02:06absolutely full of ads there's also kind
0:02:08of no guarantee that that people are
0:02:10actually really seeing it like it's
0:02:12showing some way
0:02:13err and there's kind of no relation
0:02:15between what the advertiser is paying
0:02:18and how interested the viewers are in
0:02:21the ad or whether they even saw it so of
0:02:23course this also creates high fraud
0:02:24potential you can kind of set up like
0:02:26big farms and and things to generate
0:02:29fake impressions or impressions that are
0:02:31just from BOTS or from things that will
0:02:32never lead to anything so then there's
0:02:34kind of this middle ground cost per
0:02:35click or tap CPC advertising is kind of
0:02:38in the middle it kind of splits the risk
0:02:41between publishers and advertisers you
0:02:43know the publisher is paid when they
0:02:46deliver interest to the ad their payment
0:02:48somebody clicks on the ad but then it's
0:02:50kind of up to the advertiser to turn
0:02:52that click into a sale if they if they
0:02:55can with this there is still fraud
0:02:57potential that's that's fairly fairly
0:02:59significant you know it kind of
0:03:01incentivizes publishers to place ads or
0:03:04they're gonna be accidentally clicked
0:03:06there's not a very high cost to generate
0:03:09fake clicks through BOTS and farms and
0:03:11stuff and then the third kind of ad
0:03:13payment is CPA this is cost per action
0:03:16or acquisition and that's kind of
0:03:18translated as cost per sale basically or
0:03:20cost per conversion this strongly favors
0:03:23the advertiser in the relationship this
0:03:25kind of places all the risk of whether
0:03:27the ad will work or not on the
0:03:28publishers you know the publishers are
0:03:30not gonna get paid unless somebody
0:03:32actually not only clicks on the ad but
0:03:34goes and downloads the app or creates an
0:03:36account or makes a sale there is very
0:03:39low potential for publisher fraud here I
0:03:41mean you could theoretically like and
0:03:43I'm sure people do create farms and
0:03:45people to go buy things for stolen
0:03:47credit cards or to go install apps on
0:03:49fake accounts aren't fake devices but
0:03:52this is a lot more involved than just
0:03:54like you know basic click fraud or
0:03:55impression fraud if you are if you are
0:03:57buying the ad as an advertiser you
0:03:59generally want this last method CPA
0:04:01because you're only paying if somebody
0:04:03actually buys your app or creates an
0:04:06account under site or whatever else and
0:04:07commit you know makes a sale on your
0:04:08site that's ideal for you as the
0:04:10advertiser because first of all it'll be
0:04:12easier to figure out what to pay because
0:04:14you'll know what you'll make and also
0:04:15just kind of you don't really take any
0:04:17risk there you know the risk of a
0:04:18thousand people clicking on your ad and
0:04:20then none of them buy anything you're
0:04:22not bearing that risk because you're not
0:04:23paying for people who don't buy anything
0:04:24right so
0:04:25as a buyer of ads you want CPA as a
0:04:28seller of ads whether you're running a
0:04:30website or running a store front that
0:04:32has search ads in it or if you're
0:04:34running a podcast pockets worked this
0:04:35way too as a seller of ads you want CPM
0:04:38because you want to be like well I
0:04:40delivered a bunch of people who who were
0:04:42exposed to your ad the rest is up to you
0:04:44so CPC the cost per click or tap is kind
0:04:49of the compromise nobody really loves it
0:04:50there are still problems with it but it
0:04:52is probably the most pragmatic
0:04:53compromise between them in most cases
0:04:55and then there are just a few little
0:04:57terms you might see around that are
0:04:58useful to know here eCPM which stands
0:05:01for effective CPM is when you are paying
0:05:04per click or per action you you
0:05:07basically just divide it out by the
0:05:09number of impressions so it tells you
0:05:11roughly what you are earning per
0:05:13thousand page views or impressions or
0:05:15whatever or what you're paying per
0:05:17thousand pages impressions even if
0:05:18you're not paying per impression
0:05:20technically you'll see on apples reports
0:05:22for this you'll see terms like CTR or
0:05:24TTR this is a click-through rate or tap
0:05:27through rate
0:05:28TTR is one of the last columns in the
0:05:30search ads you will see it if you do
0:05:31this that stands for tap through rate
0:05:33which is the percentage of impressions
0:05:35of the ad that result in taps on the ad
0:05:38and then see are at the very end of
0:05:40Apple's report see R stands for a
0:05:42conversion rate so this is the
0:05:43percentage of clicks on the ad that
0:05:46result in installations of the app or
0:05:49sales or whatever else that's it for the
0:05:51basic give anything to add for that no I
0:05:53mean I think the important thing I
0:05:55though to keep in mind with all of this
0:05:56stuff is that like this type of
0:05:59advertising is usually it's this like
0:06:02this is basically a numbers game where
0:06:04you the purpose of the advertising is to
0:06:07drive people to do something that the
0:06:09advertiser wants and the way you do that
0:06:12is you know you have these didn't they
0:06:14call it a funnel is the fancy term for
0:06:17it so if you ever see that's like a
0:06:18sales funnel or something that's where
0:06:19it's coming from but it's it starts off
0:06:21where you cast this very wide net which
0:06:23is where you start to get into the
0:06:24impressions a sight of thing it's like
0:06:26you just say you throw something out to
0:06:28thousands of people and then a certain
0:06:30subset of those will pay attention to
0:06:33that and a smaller subset of that of
0:06:35them will take some action as a result
0:06:37of that
0:06:38and then perhaps an even smaller set
0:06:39will actually do the the final thing
0:06:42whatever that is that's download the app
0:06:44in this case and so all of the terms and
0:06:46things are all about trying to
0:06:47understand that that process and in many
0:06:51ways it's just a numbers game at a
0:06:55certain point that obviously each person
0:06:57who's deciding whether or not to click
0:07:00on an ad is making a choice and
0:07:01obviously but from a you know the
0:07:04holistic perspective there are far too
0:07:05many people making far too many choices
0:07:07for that to be something you could
0:07:08really think about and that's why you
0:07:09tend to roll it all up into these you
0:07:12know you don't think about a impression
0:07:13you think about thousands of impressions
0:07:15gives you a sense of the kind of scale
0:07:17that this process tends to work in and
0:07:19that's why I'm gonna get advertising is
0:07:21so numbers based is just you know it's
0:07:24not really it's not that same thing of a
0:07:26sales thing where you're making a cook
0:07:28you know making a cold call to a person
0:07:30in trying to have a direct interaction
0:07:31you are just blasting out ads to all
0:07:35kinds of different people in the hopes
0:07:36that a small and a small subset of those
0:07:39people will find what you're doing
0:07:40interesting and that's why that's it's
0:07:42so data-driven and it's just I can tell
0:07:44like that's most of advertising is this
0:07:46this kind of dialing things around to
0:07:48try and hit the point where you're
0:07:51making money but not losing money or
0:07:53getting the you know if you have a
0:07:55target for the number of acquisitions
0:07:56you have you can work out well how many
0:07:58impressions do I need to generate that
0:08:00amount of interest and it's all kind of
0:08:01just you know damn multiplying and
0:08:03dividing with these kind of numbers all
0:08:05right so moving on to specifics about
0:08:08the Apple search ads first of all it's
0:08:11important to understand how the pricing
0:08:12works this is it's priced in a way
0:08:15that's known as a Vickrey auction or a
0:08:17second-price auction this is exactly the
0:08:19same way that Google AdWords works as
0:08:21far as I know or at least you know in
0:08:23the ballpark and it's also if you ever
0:08:26bought anything on eBay it is very
0:08:27similar to how standard eBay auctions
0:08:29work basically the highest bidder for a
0:08:32certain keyword or whatever the highest
0:08:33bidder for that keyword wins and gets
0:08:36their ad shown and pays but they only
0:08:38pay the amount of the second highest
0:08:41bidder on that ad so you basically bid
0:08:44what's the most you're willing to pay
0:08:46for this keyword and so if you bid $0.50
0:08:49and someone else bid 25 cents you
0:08:51we'll win the auction but you'll only
0:08:53pay 25 cents or some incremental thing
0:08:55over the 26 cents or you know whatever
0:08:56it is this actually is very favorable to
0:08:59the to the buyer to the advertiser
0:09:01because you only pay just enough to win
0:09:05the auction but no more than that so
0:09:07anything else that's between that and
0:09:09your bid you guys get to keep that money
0:09:10that that's your surplus there are some
0:09:12big limitations in the Apple search ads
0:09:15that other advertiser systems don't
0:09:18necessarily have or don't usually have
0:09:19one of the biggest ones is that you
0:09:22don't have a lot of control over the
0:09:24look and content of the ad itself the ad
0:09:28is kind of these fixed format things
0:09:30that Apple has it has your apps title it
0:09:33has your app store description does it
0:09:35does it have star ratings I think it
0:09:37does right in the example when it does
0:09:39on the in the search portal it shows the
0:09:43rating but I've seen some in the app
0:09:45store that don't and some that do so I
0:09:47don't know if it includes it or if it
0:09:49only includes it and if it has a certain
0:09:50number of ratings or if there's
0:09:52something more going on there but it may
0:09:54or may not include your star rating
0:09:56right so it's basically it basically is
0:09:59like a tiny version of your app store
0:10:01entry of your search entry so title like
0:10:04the first line of the description
0:10:06sometimes screenshots sometimes star
0:10:09ratings and one of the problems with
0:10:11this is that you as the as the
0:10:13advertiser you don't really have a good
0:10:15way to a/b test anything so like if you
0:10:17want if you want a B test two different
0:10:20description lines there and see like
0:10:22which one generates more tap throughs or
0:10:24more clicks or more sales the only way
0:10:26you can really do that is to change your
0:10:28description every so often but then like
0:10:30you can't test those at the same time
0:10:32with the same impression so it's it's a
0:10:34little it's a less scientific test it
0:10:35makes it harder to to know what is an
0:10:38effective description or whatever
0:10:40fective screenshots to show there or
0:10:41whatever else you know again you can
0:10:43test them just by changing them but then
0:10:44that changes them everywhere and you can
0:10:46only do that within a certain you know
0:10:48time granularity so it's not a great
0:10:51test yeah and I think it's also probably
0:10:53good point to point out there that what
0:10:54you're doing the what you're showing is
0:10:56the same thing that you'll show on your
0:10:58app store page
0:10:59and so you could potentially imagine
0:11:02wanting your ads to be geared towards a
0:11:06particular demographic or type of
0:11:08customer that is more specific than your
0:11:12generalized app store page which is
0:11:16something that we can't do and you know
0:11:18so you could imagine you're having a can
0:11:19add that you were running on a website
0:11:21that's geared towards a very a fan of a
0:11:23very particular thing or a particular
0:11:25style you're just try different colors
0:11:28see what people like but it whereas what
0:11:30we have here is just whatever you kind
0:11:33of have that you're doing for your
0:11:34general purpose app store description is
0:11:36the same thing that will be shown here
0:11:37and so that's you know it's nice and
0:11:41that it's less there's less work for us
0:11:43to do I suppose yeah that actually is
0:11:44really nice like I didn't have to come
0:11:46up with a creative for this I think this
0:11:48is a fancy if you ever are in marketing
0:11:50and they call it like get me your
0:11:52creative the creative ideas like the the
0:11:55picture or the text or whatever it is
0:11:57that you want to put in the ad but yeah
0:11:59we have nothing to do with that it
0:12:00should be no in many ways Apple is
0:12:01making it appear it looks just like a
0:12:04regular search result and so just
0:12:06whatever you put in there is what's
0:12:07going to show up and so your app you you
0:12:09have to make sure that you're appealing
0:12:10you know it's appealing both as a search
0:12:13ad as well as appealing as a search page
0:12:18that dholtze me that they're gonna go to
0:12:19and look at in the app store which is
0:12:21probably good practice anyway sure are
0:12:24we are sponsored this week speaking of
0:12:26ads yeah we are sponsored this week by
0:12:28Linode go to lynda.com slash radars to
0:12:32learn more sign up and take advantage of
0:12:34that $20 promotional credit with code
0:12:36radar 20 at checkout now Linode is a
0:12:40combination of high performance SSD
0:12:42Linux servers spread across eight data
0:12:44centers around the world for a fantastic
0:12:46solution to host your server
0:12:48infrastructure you can get a server up
0:12:50and running in just under a minute with
0:12:52plans starting at $10 a month that now
0:12:54gets you two gigs of RAM for 10 bucks a
0:12:57month you're able to choose your
0:12:58resource level your Linux distribution
0:13:00and your nodes location right from there
0:13:03awesome web manager portal and once
0:13:05you're up and running you can easily
0:13:06deploy boot and even resize your virtual
0:13:09server with just a few clicks
0:13:11now these servers offer
0:13:13industry-leading native SSD storage
0:13:15these are enterprise grade SSDs they
0:13:17perform crazy good these have powerful
0:13:20Intel Xeon e5 processors these are the
0:13:23fests you can get in the cloud market
0:13:25and Linode offers access to a 40 gigabit
0:13:27network with multiple levels of
0:13:29redundancy Linn oats API also exists
0:13:32this allows you to easily automate tasks
0:13:34like creating your resizing nodes or
0:13:35develop custom applications in the cloud
0:13:37and all of low notes pricing tiers
0:13:39feature hourly billing and monthly caps
0:13:41so you can pay flat per month but if you
0:13:44only use them for a certain amount of
0:13:46hours that's all you'll pay for and you
0:13:48never have to worry about your hourly
0:13:49rate going past the monthly break is
0:13:51once you reach the monthly rate just
0:13:52caps it there and you just pay whatever
0:13:53the monthly rate is this is on all plans
0:13:56and all add-on services including their
0:13:58wonderful backups and their node
0:13:59balancers which is like their fancy add
0:14:01load balancer that I use and I love it
0:14:02quite a bit anyway as it listener to
0:14:04this show if you sign up at lynda.com
0:14:06slash radar you'll not only be
0:14:09supporting us but you also got $20
0:14:10towards any Linode plan and with a 7 day
0:14:13money-back guarantee there's nothing to
0:14:16lose so go to lynda.com slash
0:14:18radar to learn more sign up and take
0:14:21advantage that $20 credit or use promo
0:14:22code radar 20 at checkout thank you so
0:14:25much to Linode for supporting this show
0:14:27all right so now that we've discussed
0:14:29kind of you know ad basics and a few
0:14:32things that make the app store ads a
0:14:34little bit different let's talk about
0:14:35the actual process of buying app store
0:14:37ads for your app the first obvious
0:14:40question here is how do you know what
0:14:43you can pay per tap so so app store ads
0:14:46are billed per tap as I mentioned
0:14:47earlier they're different different
0:14:48methods how much you can pay per tap of
0:14:52course depends on a lot of things
0:14:53what is your app cost or how do you make
0:14:55money what is the average amount of
0:14:57money that you make per new user and and
0:15:01of course then like you know how how do
0:15:03you translate between the number of taps
0:15:06on your app store page or your ad and
0:15:08how many people will actually install
0:15:11the app and make you money so this is
0:15:14obviously easiest for paid apps paid
0:15:16upfront apps where you know that every
0:15:19installation of the app will get you X
0:15:21amount of money don't forget apples
0:15:23Commission okay you know look at that
0:15:25look at you X amount of money
0:15:27and then you just have to figure out how
0:15:29many views on my App Store page does it
0:15:31usually take to make one sale and this
0:15:34is very easy to find in iTunes Connect
0:15:36analytics there's there's a bunch of
0:15:38ways to do this and people have found
0:15:40out in the various graphs and stuffs
0:15:42will be linked to that in the show notes
0:15:42was a good blog post about it basically
0:15:45paid apps is fairly easy if your app is
0:15:48not paid upfront you have to figure out
0:15:49basically what is the average amount of
0:15:51money you stand to make from somebody
0:15:53who installs your app and that is not
0:15:56necessarily straightforward depending on
0:15:57how you make your money
0:15:59it is certainly doable and you probably
0:16:01have better bracer than I do don't you
0:16:02sure I mean this is probably one of the
0:16:04harder things that I've been trying to
0:16:06work through recently is the under is
0:16:07especially for a lot of my apps that
0:16:10make their make their money from
0:16:11advertising in so it's it's this big
0:16:14video this big fantastic circle of
0:16:16advertising is enterprising to get it to
0:16:18get someone to download the app to show
0:16:19them ads to give me money so that I can
0:16:21buy more ads and you know that the cycle
0:16:23continues but for them it's much more
0:16:25complicated because yes there's no
0:16:27upfront one-time purchase it's you know
0:16:30that person who you know I'm paying for
0:16:33a tap a certain number of people who tap
0:16:35will then install the app they will then
0:16:38use the app at a particular for a
0:16:40particular a number of you know I'm out
0:16:41of time which is if you're looking at
0:16:44your analytics you're looking for
0:16:45something called retention which is
0:16:47showing you you know roughly how what
0:16:50percentage of users will continue to use
0:16:52your app after a particular time you
0:16:54know so if you had say a 30 percent
0:16:57retention after 30 days that means 30
0:16:59days after someone downloads it there's
0:17:00a good chance that they are continuing
0:17:02to use it so you can kind of get a sense
0:17:04of how often they'll use it then you
0:17:06have to work out how many sessions they
0:17:08typically do per day of those sessions
0:17:11how many impressions you're showing on
0:17:12your ad network and then you work out
0:17:15that from there how much revenue you get
0:17:16per impression and so on and so you it's
0:17:19much more inaccurate and you know I've
0:17:22been trying as best I could over the
0:17:24last you know a couple weeks as such as
0:17:27became a thing to try and really get a
0:17:28good number for my apps but at this
0:17:30point I just kind of have a vague
0:17:32estimate of like it costs I get maybe
0:17:34this much per user and I'm just kind of
0:17:37basing my advertising on that because I
0:17:39think it's also probably worth pointing
0:17:40out
0:17:41well it's the most straightforward way
0:17:43to price this is just in the sense of
0:17:44you know if I pay a penny less then I
0:17:48make from that user for the ad then I'm
0:17:52making money but you could also just
0:17:55view it as a marketing expense that
0:17:56certain kinds of marketing you're just
0:17:58trying to get people interested in your
0:18:00application and exposed to it and kind
0:18:03of it's more I guess what you'd call
0:18:04brand advertising that you're I'm glad
0:18:07we're you we can worry less about it and
0:18:09you know and certainly some of my
0:18:11experience early on that's kind of what
0:18:12I'm doing I'm just kind of curious to
0:18:14see how inexperienced what it costs to
0:18:16get users and thinking in some ways that
0:18:19you know having my app in more people's
0:18:22hands will have a bigger impact down the
0:18:25road then they're not and it may not
0:18:27necessarily match up one-to-one in terms
0:18:29of you know I paid 25 cents I made 26
0:18:33cents so I'm good it may be that I'm
0:18:36paying overpaying by you know 10 cents a
0:18:39user which isn't sustainable necessarily
0:18:41in a direct sense but it may be if that
0:18:44those extra downloads are broadly make
0:18:47you know helping your app you know from
0:18:50an exposure perspective so it's Tiffany
0:18:52it's really tricky thing to work through
0:18:54but you know you just kind of have to
0:18:55sit down in your analytics whether
0:18:57that's you know something like fabric
0:18:59that you're using which is what I use
0:19:00and I think what you use is or iTunes
0:19:03Connect analytics can do a lot of this
0:19:04for you as well
0:19:05and you just kind of have to sit in you
0:19:07take the numbers put it into Excel and
0:19:09you do some crunching to see what makes
0:19:11sense for you yeah I mean you know thing
0:19:13to think about is like how what
0:19:14percentage of your new users come
0:19:16through clicking adds because like you
0:19:18know if you're thinking that like you
0:19:21make 25 cents per user but you know 10
0:19:25percent of your users come from clicking
0:19:27ads and the rest come from other ways
0:19:28then like you actually might be able to
0:19:30bid a lot more than what that is to get
0:19:32those users you know and again it
0:19:33depends on like what are your priorities
0:19:35are your priorities to get more users to
0:19:37spread it wider and then you know to me
0:19:39maybe reap the rewards of that down the
0:19:40road with just a wider bigger user base
0:19:42and word-of-mouth marketing like there
0:19:44or you know or do you really need to
0:19:45care about like absolutely minimizing
0:19:48cost of acquiring these users to begin
0:19:49with and there's every app will be
0:19:51different in this way what you know your
0:19:52needs your priorities these will all be
0:19:54different
0:19:54and it really depends a lot on your app
0:19:56and how much you stand to make per user
0:19:58really and this is actually one thing
0:20:00that I think this is actually helpful in
0:20:02this new environment because III think
0:20:04this will make people think about their
0:20:06business models a little bit more and
0:20:07maybe charge more money and charge
0:20:09higher prices for apps that really that
0:20:11that that can earn it that deserved the
0:20:13higher prices so I think this actually
0:20:15might help raise App Store prices as
0:20:17people figure out you know it would be
0:20:19really a lot better if I could bid on
0:20:20these keywords and be a little more
0:20:22competitive on the ad side but in order
0:20:25to do that I have to charge a real price
0:20:26for my app or I have to have some kind
0:20:28of recurring revenue scheme or something
0:20:29like that so ultimately I think this
0:20:31might actually help ad pricing
0:20:32substantially yeah or at the very least
0:20:34it'll help I think it's definitely been
0:20:36a constructive exercise for myself to
0:20:38have to try and put a dollar amount on
0:20:41each user yeah which it for someone who
0:20:44loves numbers as much as I do is
0:20:45something that I've never really done
0:20:47and so it's definitely been a
0:20:48constructive thing - if forcing you to
0:20:51think down sit down and actually be like
0:20:53okay what is a customer worth to me and
0:20:57to put that into a number because even
0:20:59beyond advertising that's a very useful
0:21:01and important thing I think to
0:21:03understand because even when you're
0:21:04thinking about things like customer
0:21:06support or things like other support
0:21:10expenses or server hosting or things
0:21:11where you if you can put your a number
0:21:14on what a user is worth monetarily to
0:21:17you then you can probably make more
0:21:19informed decisions about the other
0:21:20things that are involved in that are
0:21:22costs for running an app yeah all right
0:21:25and then I figure we spent our last few
0:21:26minutes talking about keywords you know
0:21:28what keywords do you bid on how you know
0:21:30what do you buy what what do you bid
0:21:32against apples and apples implementation
0:21:34here is a little bit restrictive in that
0:21:36they even mentioned this during the
0:21:38during the the introduction of this that
0:21:41if Apple does not deem the keywords that
0:21:43are being searched for relevant to your
0:21:46app and that's relevant by whatever
0:21:48internal standards they're using that's
0:21:50kind of a black box to us but basically
0:21:52if Apple does not deem keywords relevant
0:21:53to your app your ad will not be shown
0:21:55against those keywords no matter what
0:21:56you're going to bid on it which is
0:21:58interesting for example like one of the
0:22:01things I decided a bid on an overcast
0:22:03sad campaign was some terms related to
0:22:05the word
0:22:06I do because podcasts are kind of like
0:22:08listening to the radio and so I want to
0:22:10kind of get those people to and all of
0:22:13those if you look at their horse those
0:22:14all have zero impressions and so it
0:22:16seems obvious that Apple does not deem
0:22:18the word radio relevant to my app and so
0:22:22even if I'm willing to pay that's I
0:22:24can't get those key words and this you
0:22:27know so this is this is actually this is
0:22:29probably a good thing so what it means
0:22:30is like I couldn't bid against something
0:22:33like Twitter or Instagram for overcast
0:22:35to try to you know cash in on big search
0:22:37terms that aren't actually relevant to
0:22:38my app at all but also that means that
0:22:40all of us small indie developers we
0:22:42can't all be outbid in every single
0:22:44category by things like clash of clans
0:22:47and like casinos like apps that make
0:22:48tons of money per user to the point
0:22:50where they're willing to spend whatever
0:22:51whatever it takes to even get users who
0:22:53are searching for podcast apps because
0:22:54they don't care it's it's a user as a
0:22:56user to them and they they make so much
0:22:57money they can do that so this basically
0:22:59makes that impossible and it kind of it
0:23:01kind of gives people a more fair chance
0:23:03of their ads being shown in categories
0:23:06that are not that hotly contested for
0:23:08things like you know high profit games
0:23:09if you actually make a game well that
0:23:12might be harder but I'm sorry about that
0:23:14I don't any harder for you anyway though
0:23:16that's actually true your upside is
0:23:21generally higher but it's harder to get
0:23:23it anyway there's also this big question
0:23:26about whether to bid on your competitors
0:23:29app names and and we've had some
0:23:31discussions on Twitter here and there
0:23:32about this but my my position on whether
0:23:35I've been in competitors names is yes
0:23:38you should do it and that's just part of
0:23:39the game and your competitors are gonna
0:23:40bet in your name too and that's just
0:23:42part of the system and that in general
0:23:44it not only should this be considered
0:23:46like you know unnecessary side of this
0:23:48but I wouldn't even call it like a
0:23:49necessary evil I'd call just a feature
0:23:51of his system because you know you're
0:23:53thinking you know if you're offended by
0:23:55the idea of competitors bidding on you
0:23:57you're probably thinking of bigger
0:23:59competitors bidding on your apps name
0:24:01but if you think about it the other way
0:24:03you can bid on bigger apps names and
0:24:06that is more powerful because that you
0:24:08think like you know like one of the
0:24:09biggest podcast apps out there is
0:24:10stitcher right and and I'm I bid on
0:24:13stitchers name B and I've gotten click
0:24:14throughs and installs based on that
0:24:15because I even if somebody's not
0:24:18searching for overcast
0:24:19irrelevant it's it's a perfectly valid
0:24:21alternative in my opinion it's a lot
0:24:23better alternative obviously and so I
0:24:26can get users that are searching for my
0:24:28competitors apps who are bigger than me
0:24:30and that is that is a very kind of
0:24:32freeing opportunity that I have it like
0:24:34that that's a grab from you take
0:24:35advantage of so if you think about it in
0:24:37that direction I think it makes a lot of
0:24:39sense why they allow that and that's
0:24:41that's just part of that like all
0:24:43advertising kind of works that way
0:24:44that's I think that's part of the game
0:24:45and I think that's a fair game and if
0:24:47you decide that you don't find that
0:24:50ethical that's fine you don't have to do
0:24:52it but but i think i think standard
0:24:55practice is going to be almost
0:24:57everyone's going to do that so if you
0:24:58don't do it you're probably gonna miss
0:25:00out on some some pretty good sales
0:25:04opportunities there sure I mean it's
0:25:06also probably worth saying that the way
0:25:08Apple's system works and this is
0:25:10actually to your point about relevancy
0:25:12if you're curious what Apple thinks is
0:25:13relevant to your application the easiest
0:25:16way to do it once you have been running
0:25:17a campaign for a little while is to just
0:25:19go into there the you sort of if you tap
0:25:22down a little bit into your campaign
0:25:23you'll get the breakdown of where what
0:25:26your impressions are by keyword and many
0:25:29of those keywords won't be things that
0:25:30you put in they'll just be terms that
0:25:33Apple thinks are relevant and so at a
0:25:35high level I like I'm in my mind if it's
0:25:38on one of those if it's in that list
0:25:39that Apple is doing they're automatic
0:25:43matching with then that's probably
0:25:45roughly the universe of what Apple
0:25:47thinks is relevant to me so if you're
0:25:50kind of curious with that relevancy is
0:25:52and the reality is in that list with me
0:25:55putting nothing into it I'm seeing the
0:25:57names of my competitors apps because
0:25:59Apple knows that they're very relevant
0:26:01like the in many ways the most relevant
0:26:04results are when someone is very
0:26:06specifically looking for an application
0:26:08that is very similar to yours and I
0:26:11think like you said it's just that is
0:26:13just the nature of the app stores that
0:26:15it's very competitive it's very um I
0:26:18wouldn't cut throats the wrong word but
0:26:20like that's the reality is that we were
0:26:21you know we were competing for a very
0:26:23limited number of of users attention and
0:26:26so if it's like just not doing it like
0:26:30not then just like don't buy ads be
0:26:33and don't go down sort of into this road
0:26:35because the nature of what we're doing
0:26:37here is we're just paying money to get
0:26:39Ateneo to pull the attention away from
0:26:41something else because whatever was
0:26:42gonna be the organic first hit there you
0:26:45know this is what we are now pulling
0:26:47money away from by buying it and that's
0:26:49maybe kind of weird but that's the
0:26:51reality that's the system that Apple's
0:26:52building and if this is the marketplace
0:26:55in which we want to work and operate
0:26:56then I think taking full advantage of
0:26:58that as best we can is serve the
0:27:00responsible appropriate thing for us to
0:27:02do yeah exactly
0:27:04and I would say like the the biggest
0:27:07takeaway I have is you know you might
0:27:09have some idea of what you can pay what
0:27:11things will cost how effective ads will
0:27:13be but nothing beats actually trying it
0:27:16and Apple made it is really easy on us
0:27:18because they sent everybody $100 promo
0:27:21credit and I don't know if all new
0:27:22accounts will get that or if only the
0:27:24existing dev accounts got it last week
0:27:25but the hundred promo credit is awesome
0:27:28because you couldn't you use get $100
0:27:30worth of free advertising and that can
0:27:31be your cap and that can be it so you
0:27:33couldn't literally get free ads and then
0:27:35learn the data from those ads you can
0:27:37learn what keywords are working you can
0:27:39learn what your average cost per tap or
0:27:41your average cost per installation ends
0:27:43up being and that can further inform
0:27:45what you do if you don't get the free
0:27:47promo credit or if you want to go past
0:27:48that you can actually learn a lot with
0:27:50not that much money relatively speaking
0:27:52to the cost of computers and developer
0:27:54accounts and developing apps and stuff
0:27:56like if you want to put in 50 bucks a
0:27:58hundred bucks that seems like a lot of
0:28:00money into a lot of people but if you
0:28:02put that in there you will get so much
0:28:04data out and you will find out like
0:28:06really what these things are what these
0:28:08things do for you what they cost and
0:28:09that's true of all advertising like
0:28:11nothing beats just trying it and a lot
0:28:14of times like you know a lot of times
0:28:15you won't learn you can theorize as much
0:28:17as you want you can you can run
0:28:19theoretical numbers but you have to just
0:28:21try it and sometimes it doesn't work out
0:28:23and that's kind of the risk you take by
0:28:24trying it but usually it and if you can
0:28:27make enough money per user of your app
0:28:29to to cover what these things tend to
0:28:32cost and like it like my my cost per tap
0:28:35is in like the 20 cent range right now
0:28:37and it's going down as everyone's promo
0:28:39credits run out so to give you some idea
0:28:41what you might be paying what might be
0:28:43spending here
0:28:44you know if you can figure out how to
0:28:47make that work for your app just try it
0:28:49and and just and see what it does for
0:28:51you
0:28:52this is now just part of the it part of
0:28:55the reality of working in the App Store
0:28:56and so you got to try it put some money
0:28:58into it and understand that sometimes
0:29:00it's like I've put our money into
0:29:01advertising and it kind of feels I'm
0:29:03just throwing money away and in some
0:29:05ways maybe that's the case I'm not doing
0:29:07it right but the reality is just the
0:29:09cost of doing business and you owe it to
0:29:11yourself if you take your business
0:29:12seriously to at least try it I think and
0:29:15understand it rather than just just
0:29:17counting it out of out of hand
0:29:19exactly all right well we're out of time
0:29:22this week thank you everybody for
0:29:23listening and we'll talk to you next
0:29:24week bye