Codestrong 2012 keynote how to build a top ten app

Editor's Notes

• #3: You know who I am…WHO ARE ALL OF YOU?
• #6: The situation is very similar on the Android side of the house
• #7: Marketing is not a magic bullet that will automatically result in tons of sales. The most important thing is to develop a good app. Marketing leverages your awesome app for (hopefully) awesome results. Here's the truth: Technical execution is relatively trivial (besides unique, well-researched and tested algorithms); it is the marketing execution that sets the hot products apart from those languishing at #1,380 in a given category.
• #9: So, what is a good app?That's really up to you and what your audience demands. Clear is a great example of this: in a super crowded to-do list space, they focused on unique UX and it served them well. Here's a good exercise to figure out if you have a unique concept: can you pitch your app in 80 characters or less. That's about the size of a standard email status.
• #11: Let's set some expectations and some context. First of all, we need a business plan if we're going to be building a business. And the biggest item on a business plan: REVENUE. There are only 4 ways to make money directlyfrom your mobile appsI’ll bet a lot of you are making money with contracting, which is awesome! But the scope of this talk will cover how to make money from your apps
• #12: Unfortunately, without any sort of understanding of how popular your app will be, you can't really build an accurate projection of revenue.
• #13: Unfortunately, without any sort of understanding of how popular your app will be, you can't really build an accurate projection of revenue.
• #14: Test, test, test.Fortunately, we've built a little tool to help you set some realistic expectations when it comes to revenue, at least from in-app advertising. Use our Developer Economic Tookit to get a good handle on what you can be expecting.The best plan of action is to see what works best and delivers the highest ROI for your business.
• #15: Monetize from the get-go
• #16: What we're really trying to understand is the Lifetime Value (LTV) of each customer. Keep in mind, right now we're just working with projections. There is no way to actually know these numbers until the app is out in the wild. Once you do have some data, however, this formula will help you understand the LTV of each of your users:
• #17: Unfortunately, all too often we focus on Step 2: Coding to the detriment of the other stages.Marketing applies to every aspect of the life cycle, not just after your app has gone live in the app store. Marketing even has a role in planning/ideation. Make sure to do some market research, not only target market feedback, but also take a look at the app stores and the apps that are similar. The App Store is your best comp intel tool with 700K+ of your competitors in a single catalog. Pay attention to the highest rated, and make sure to note what their users appreciate. Oh, and don't forget to remember what they hate.Ok, so when do I start marketing? Some have taken the approach of marketing the day you start coding, and that seems to have mixed results. The benefits seem to be that you can maybe generate some early buzz. Of course you have to weigh that against a realistic timeline of your app.
• #18: Anybody know these titles? Both were hyped up YEARRRS before their launch (DNF hasn’t even launched yet), and pissed off all the fans and gamers.
• #19: If Bear can’t survive, how will your users???
• #20: But as long as you don't plan to start marketing YEARS before you launch and condemning your app to development hell, generating early buzz has never been easier. You can throw up a LaunchRock page, create a Twitter account, even perhaps get a Tumblr or Wordpress blog going. If you have a little more ambition, you can easily make a teaser trailer for your app with something like Screenflow or iMovie. Video humanizes digital communication and can spark interest in your app as consumers will get a better feel for what they can expect. Look at it this way, launch day is a point on your marketing timeline, not Day 1. Use every day to build up your list of emails and Twitter followers so that your launch day activities have more reach.You can also use the promo codes in the app stores effectively. The impact is lessened for the Android platform since Google’s approval time is so short, but on iOS you can leverage the long approval time to your advantage. You can give people access to the app before it is launched, but after it is approved. The temptation is to click that green button as soon as you’re approved from Apple. But by utilizing the 50 promo codes that Apple gives you to get some initial feedback and make sure the app runs well can be very valuable. And unlike the hassle of beta testing (commitment, Test Flight and provisioning), giving promo codes not only incentivizes your friends to actually download your app, but it is doable with a single code in the native App Store.In fact, you can leverage the promo codes very effectively by reaching out to bloggers of all shapes and audiences and offering an early peak. There are legions of bloggers out there from the “tech press” to mommy bloggers who rate and review apps for their readers.
• #21: Of course, if you're considering doing some beta testing (which is highly recommended for not only errors but UX/UI feedback), these early buzz generators will help you in acquiring a cohort of trusted testers who can help you polish your app. And in this age of SaaS for everything, there are services that can help you with beta testing, such as Test Flight.
• #22: But while you're coding, make sure to consider elements that will help you with marketing later on:
• #23: Its easy to get bogged down in coding and making sure that all the code is as perfect as possible; so remember - the user never knows how pretty your code is. At the end of the day, its all about shipping. So code and code and code, until you have a minimum viable productand then ship that out to your beta testers. Get some feedback, fix some critical errors and make a decision to push to the App Store or not.All the while, keep generating some early buzz. Keep that Twitter account active and keep doing market research to make sure you're not putting three sunless and friendless months into a product that fills no need.
• #24: I've coded, and I'm about to launch. In fact, my app is already live and has approval. What should I do?Now comes the really fun part: deployment and launch day! Push your app live to the App Store, pop open a few beers and wait for the money to start pouring in.
• #26: Remember, launch day builds on top of all the pre-launch marketing that you've been doing, so the more successful your buzz generation are, the easier it will be to have a successful launch. But it is not going to happen over night. In fact, with close to a million apps, it is going to be hard to get noticed. This is where all those APIs you considered and built into your app can come in really handy.But don't count on just a social integration to help your app reach stardom. It can really help, as it did with Turntable.fm and as Pinterest, but again, these sites had a really compelling, and at least somewhat unique value proposition. You can leverage your users' social graphs to gain more traction, but be wary of acting in a spammy kind of way. You don’t want to be Socialcam.But don’t put all your eggs in the social basket. At Tout, we tried to leverage OpenGraph to drive sign ups and sharing activity, and the lack of results plus the fact that it was bothering our users to have to authenticate all the time made us pull it out.In the end it is still about the perceived value of your app to a potential user
• #27: In practice, not many outside of the app marketplace providers have a complete understanding of the algorithms, but we can surmise a few things. My counter part in London, Terence Eden actually got up close and personal with the Android team and learned a few things. Check out his blog post here, but of the more than 200 metrics that play a part in Google Play rankings, the big three are:Ratings, length on handset and velocity. Remember that the ratings are an average and not an absolute.There are similar guidelines for both the Apple App Store as well as the Amazon Appstore. So far we have stressed the importance of earning and acquiring ratings by making high quality apps, soliciting user feedback, and listening to it. All of these actions lead not only to better ratings but a longer length on the handset because you have happier users.
• #29: Make sure to have visually appealing assetsThis is a pretty big deal as your app icon will be the first (and perhaps only) interaction that potential customers will have with your product. Of course, your screenshots are also visual assets and should up to snuff.This change is even more apparent for the new App Store on iOS. Think of your logo as your storefront, as users are trying to determine which apps will get their time, visually appealing icons matter.Also, do your research. I was searching for a fashion app the other day and of the top 10 results, 9 had black icons with white text. This would have been a really easy opportunity for one of the apps to stand out from the rest and really attract users.
• #30: http://www.dwellable.com/blog/How-We-Doubled-Our-Android-Install-Rate-in-One-Hour
• #31: http://www.dwellable.com/blog/How-We-Doubled-Our-Android-Install-Rate-in-One-Hour
• #32: Make sure the app description is bulleted and frequently updated.Simple bullet points that call out key features are an effective and easy to way to explain the game. Frequent updates demonstrate that the app is actively maintained and helps users steer clear of apps that are not well supported. Make sure to also include any awards you've won hereNotice that you barely have any room for characters before you’re cut off by the “More” in both the “description” and “what’s new”
• #33: Make sure to cross promoteReal estate on the App Store isn't cheap, so don't take it for granted! Leverage what success you do have to increase sales across your portfolio by cross promoting your apps on each other's app store landing pages.There are some very good arguments to just pushing a number of apps to the app store so that you have a larger customer base to cross-promote to
• #34: Make sure to have ratings (and high ones at that!)With so much choice in the app stores, social proof becomes more and more important. The main avenue of social proof in the app stores is the number of ratings and stars that a user sees associated with your app. This serves as a bit of a feedback loop as well, considering you can't have high ratings without making a high quality app and listening to your users.--If you were starting a traditional, brick-and-mortar business the app store landing page would be analogous to your store front. In fact, its your distributed store front. You have customers standing right there in front of you, you just have to entice them to walk up to the door and open it. Aka put in their password.Ok, so you have a killer store front/app landing page. Its got thousands of 5 star ratings, you have awesome graphics and just won a whole boatload of awards. But you still need people to find it, to walk past the store front. And here's where it starts to involve some dark magic: app store rankings.
• #35: …And of course getting noticed.
• #37: Great so now you have a high converting landing page. Let’s step backwards up the user acquisition funnel, how are users going to find your app?
• #38: Vyclone’s technique of using a glowing review from TechCrunch provides some seriously good validation for the app. This can serve to intrigue users before they’re fully aware of what your app’s value add is.
• #39: Of course, not everything is on you the developer, some of this relies on the platform. For instance, Apple is really pushing “Genius” as an alternative to categories and search. And it makes sense, as the app catalog grows and grows, categories become less and less effective. Ian Sefferman of MobileDevHQ, an ASO SaSS tool, makes a great point that its similar to 1996 on the web when Yahoo categories and dmoz had really run their course and the discovery mechanism shifted towards search. Yahoo & Altavista to Google.But Genius isn’t that good yet, and in my opinion, even iTunes genius needs a lot of work. As you can see here, its recommending a “Productivity” app to me based on the fact that I have the iAds gallery app on my phone. Except this app has to do with calendars and scheduling, not quite related.
• #40: In practice, not many outside of the app marketplace providers have a complete understanding of the algorithms, but we can surmise a few things. My former colleague, Terence Eden actually got up close and personal with the Android team and learned a few things. Of the more than 200 metrics that play a part in Google Play rankings, the big three are:Ratings, length on handset and velocity. Remember that the ratings are an average and not an absolute.There are similar guidelines for both the Apple App Store as well as the Amazon Appstore. So far we have stressed the importance of earning and acquiring ratings by making high quality apps, soliciting user feedback, and listening to it. All of these actions lead not only to better ratings but a longer length on the handset because you have happier users.
• #41: In practice, not many outside of the app marketplace providers have a complete understanding of the algorithms, but we can surmise a few things. My former colleague, Terence Eden actually got up close and personal with the Android team and learned a few things. Of the more than 200 metrics that play a part in Google Play rankings, the big three are:Ratings, length on handset and velocity. Remember that the ratings are an average and not an absolute.There are similar guidelines for both the Apple App Store as well as the Amazon Appstore. So far we have stressed the importance of earning and acquiring ratings by making high quality apps, soliciting user feedback, and listening to it. All of these actions lead not only to better ratings but a longer length on the handset because you have happier users.
• #42: Explain what the velocity means in this contextGreat, well how do we affect that last major metric: velocity of downloads. Its actually significantly easier than anything you've done up till now; way easier than designing and building a high quality app.
• #43: You just have to pay money.Its that easy. There are a number of different options for what's known as performance advertising. Basically, that means you pay on a cost-per-download or cost-per-install basis. The cost-per-download will vary drastically from country to country, from genre to genre and from network to network. This is why its good to promote your app on multiple ad networks. And here's the super awesome conclusion:If your cost per acquisition is lower than the LTV of your user, your performance campaign has an essentially limitless budget.Remember, we wanted downloads to not only acquire users, but to affect the velocity of the app in the app store as well. For this, burst campaigns are quite effective. A burst campaign is a campaign over a few days, usually a weekend, that drives a tremendous amount of downloads in a short period of time to dramatically increase the app’s ranking in the app store.If the app breaks into the top 10 or 20 in a given category, then users can start to organically find it on the app store's main pages. It is not a coincidence that once an app breaks into the top 10, it generally stays there for some time.
• #44: Great, we've applied our business thinking to steps 1 through 3 of the app development lifecycle. But remember all those annoying analytics APIs that we had put in way back when? Oh and those error reporting SDKs? Those are all going to become super useful right about... now.Analytics can start to help you understand how your users are using (or not using) your app. You might have thought that the killer feature was a certain view or screen but it turns out that users actually hate it and do everything they can to avoid it. The only way to know this in any meaningful way is to have analytics installed from the get-go and continually analyze the usage patterns of your users.Errors are a serious cost to your business. Every error is a lost in-app purchase opportunity, potentially a lost user and almost certainly a less favorable review. Make sure you're aware of the errors and crashes that your users are experiencing by using something like Crittercism or Bugsense. So take all this new data and feedback that you've got and start the whole cycle over again. Of course, except starting with an already live (and hopefully already popular) app. Plan out some features and bug fixes, code them up, push it live and around and around it goes. Of course, you can use these updates strategically and space them out every 2 or 3 weeks to keep users coming back to your app, and new users downloading it.Cool little trick I’ve been noticing more and more in apps is to reward repeat users after a crash. For instance, if the app crashes and that same user opens it again (from what I’ve seen with no time bounds), the developer rewards the user with one of the smaller in-app purchase goods for free. I think this is a great way to generate good will, feedback and repeat usage of your app even if it crashes.
• #45: First 200 developers to sign up on InMobi get $100 in ad credit