Here are my notes from the Mobile Mobile Conf that took place in Kraków this week. I just wrote down what I heard, so it might not be completely correct, but I figured it will still be useful.
The Future of Mobile Starts Now (@sascha_p)
- What’s coming:
- smart watches (“they’re terrible”)
- Google Glass
- apps for car computers (the car industry is slow in adopting new technologies, Ford is leading in this regard)
- perceptual computing – controlled by eye/hand tracking
- Low cost tablets are the next big thing – they’re selling in Asia right now in tens of millions for $100, they will be many people’s first computer. There are countless producers you haven’t heard about selling countless models, some of which sell as many devices as Apple.
- Smartphone and tablet OS market share in Asia is around 85-95% Android, only 4-5% iOS (but even if iOS global market share will be 10-15%, it won’t be a problem for Apple).
Technology – The Power and the Promise (@USA2DAY)
- When it comes to accessibility, iOS rules, Android is still catching up. You can do everything on the phone using voiceover, even find streets on a map (in Apple Maps).
- Interesting apps/devices:
- Money Reader – scans banknotes using camera and tells you what it is
- Fleksy – easy typing on the on-screen keyboard with auto-correction and voice help (also useful for non-disabled people)
- Apple FaceTime – can be used by deaf people for speaking using sign language
- FaceXpress – detects emotions of a person seen on a picture/video (can be useful for people with autism)
- Google Glass – it will be an amazing help for blind people
- Autonomos Labs – driverless taxi ordered from a mobile
- You can be “temporarily disabled”, e.g. when you’re driving and can’t touch or look at the screen, so every app or feature useful for disabled people can be potentially useful for anyone.
- Thanks to technology some disabled people can now live a normal life they couldn’t live otherwise.
The Elements of a Killer Idea (@jerols)
- “Ideas don’t matter, it’s all about execution” – not really, ideas matter too. People won’t buy an awesomely implemented but useless app. Idea by itself doesn’t matter a lot, but it’s the foundation of your app.
- “Languages” made as much money in one day as “Grades” made in 2 years – the difference was the idea.
- What can YOU make better than anyone else? What knowledge, experience, resources, connections you have that others don’t? Can you build it for yourself, to solve your own problem?
- A good app has a clear message and purpose – you should be able to explain your app’s purpose in one sentence. How are your users going to describe it when they share it with others? Why is it valuable to people? This helps you focus.
- Design also matters. It’s hard to get noticed, so micro-ideas are important (how interface works, how easy the app is to use).
- “Don’t make me think”
- think like a human, use the mental model of the user, not your app or database
- don’t make me work (type too much, sign up before use) – users need to understand the value before doing it
- do usability testing, you have no excuse not to do it – you’ll find tons of things you haven’t thought about
- make the app beautiful, make it enjoyable to use (don’t go overboard though)
- touch is magic, it gives you an illusion you’re manipulating real objects
- you need to go the extra mile, take care of all the small details, animations etc.
- “The two App Stores”
- hit apps: cheap, wide appeal (simple todo list, messaging, translation, games etc.)
- has to get to the top of the list, makes money on the volume
- if you want to make a hit, rely on charts, not on search (not enough people use search)
- you must try to get featured on the App Store by Apple
- premium apps: limited appeal, niche, high price (e.g. some more complex productivity apps)
- needs to provide a lot of value
- not impulse downloads (people do a lot of research before they buy)
- rely on search SEO, reviews on app review sites etc.
- premium apps with wide appeal: really hard to get there
- most apps: niche idea with low price, doesn’t make a lot of money; developer comes up with a common idea that people won’t pay a lot of money for, but the app doesn’t get really popular
- hit apps: cheap, wide appeal (simple todo list, messaging, translation, games etc.)
- Find a vacuum: find an existing app that is popular but is missing something important (e.g. when Siri was released everyone was writing “like Siri but …” apps)
- Browse the App Store, see what’s popular right now, what are the trends, what your competitors are doing.
- Mobile app stores:
- iOS App Store
- Android (Google Play)
- Amazon Appstore
- Barnes&Noble (NOOK Apps) – smaller, but people are used to spending money
- Windows Phone Store, BlackBerry World – probably won’t take off, “the train has left the station”
- Rovio made 50 apps that weren’t a success until they made Angry Birds.
- There are ~800,000 apps on every app store, so there’s a pretty good chance there’s already an app that does what you want to do, even for niche apps, and lots of them are free (especially on Android).
- Once you’re at the top, it’s easy to fall quickly.
- 60% of apps don’t break even, less than 12% earn more than $50K.
- The bar is set high so aim high, you can’t aim for the middle. You have to make a lot of effort to make the app perfect. Awesome UI and UX means everything, you have to spend a lot of time on this.
- Marketing is important; top apps spend 14% of time on average on marketing.
- Getting on the top list is important. Also, if you get there before Christmas, you stay there for a week or two because the list is frozen (and it’s a period when a lot of people get new devices).
- Avoid those categories where there’s a lot of competition.
- Making the app cheaper doesn’t always mean it will be downloaded more, it’s not as simple, so you might make less if you lower the price. Check the statistics.
- Freemium makes a lot of money now, but it’s not always the right answer. People still pay for apps, subscription might be a good idea too depending on the type of the app.
- App stores don’t need to be your only channel, you can try to sell the app directly too (if the device/OS allows it).
Product is the Byproduct (@holman)
- To build a great product, build a great culture.
- Don’t just hire all great developers, hire people you’d like to work with. You want fixers, people who get irritated quickly and start fixing things.
- “There are thousands of developers who are extremely smart and talented. All of whom I would never, ever hire here.” – @kneath
- “Any time you interview a potential hire, you need to ask yourself not only if they’re talented or collaborative but also if they’re capable of literally running this company, because they will be.” – Valve
- Diverse groups are more creative than homogenous groups.
- Be flexible (location, hours, workload etc.).
- Intel Macs were initially developed by one Apple engineer that went to the east coast and worked remotely because of family matters. If Apple didn’t let him do that, it’s possible that we wouldn’t have Intel Macs now, and we probably wouldn’t have iPhones either (or Androids).
- 66% of GitHub developers work remotely. Everything is asynchronous, so people can choose the hours they want.
- Be genuinely excited as a company about the things you work on, blog about any small improvements etc. Make some noise and buzz, make people excited before the launch.
- “People on the internet are dicks” – you should try not to care about them. Care about your fans, people who actually use your product, but not about haters and skeptics who might use it if you add something but probably won’t anyway (be wary of “if”).
- Nothing great was ever not shipped.
Marketing Isn’t Bullshit (@janaboruta)
- “The product should market itself” – no it won’t, you have too much competition.
- Engage your current users if you have them (e.g. invites for webapp users to the new iPhone app – they sent presents to their top 50 users with invites in the box).
- Send emails to bloggers, app review sites etc.
- Avoid PR companies, create relationships with the reviewers yourself.
- Treat them like humans, they’re people like us, just with more influence. Keep the emails short, skip buzzwords and bullshit.
- Show sneak peeks on Dribble.
- Set a launch date and goals for the launch, have a plan for the launch prepared earlier.
Your Idea Sucks (@MatManferdini)
- Don’t make an app in an area you know nothing about.
- Don’t make an app in a technology you don’t like.
- Don’t sign any NDAs, especially before you actually hear the idea.
- Fuck glory, don’t be driven by the idea of startup glory. 90% of startups fail.
Going Big with Mobile Apps (@patrickbroman)
- In 2013 it’s mobile or die. But things always change, so if you aren’t fast enough, you’ll die anyway.
- (there was a story about how Spotify’s code architecture evolved, but I didn’t get all the details)
Sensors and Sensibility (@briansuda)
- Humans have actually more than 5 senses – there’s a sense of temperature, acceleration, time…
- Animals can also detect e.g. electricity, magnetism, pressure, current, polarized light…
- Phones have most of that, and some other things (GPS, light, tilt, WiFi, water, infrared, bluetooth, sound, touchscreen…)
- In future: NFC, radiation, temperature, barometer, humidity, pressure…
- cubesensors.com, mylapka.com – various external sensors that can be connected to an iPhone
Tap to Write History: Our Digital Legacy (@_aitor)
- The new gods are asking for sacrifices like the old ones, and the thing to sacrifice this time is our privacy.
- Evernote promised their users their data will be preserved for 100 years, even if the company is closed.
Ubiquitously Connected to the World (@soffes)
- You can buy a lot of Wifi/bluetooth connected things: light bulbs, door locks, heaters, etc. You can e.g. turn off the lights from the terminal, or turn them to red when there’s an error on production.
Design for Ecosystems Instead of Screens and Pages (@shoobe01)
- Not everyone has a smartphone. Facebook for old Nokia feature phones – 80 mln active users.
- Users might not care that the network is down now, don’t disturb them with a huge error message that blocks everything. Connection problems are inherent in wireless communications, if your app can’t handle that gracefully, people won’t use it.
- If you don’t do it right, people will notice. People have an intolerance for bad experiences.
- Facebook started to get 4x as many users immediately after they switched from the hybrid HTML5 app to the fully native app.
- Design device-agnostic services, don’t design for specific devices or screens.
- Don’t assume your users will use your site or app in some specific way (e.g. that they will start from the home page).
- Don’t throw away half of the features and content when designing mobile version of the site just because the screen is smaller.
Building Apps for Emerging Markets: Fact and Fiction (@furiousleddy)
- 70% of mobile projects fail.
- It’s easy to make stupid mistakes if you don’t consider the context (e.g. including QR codes to download an app in an in-flight magazine).
- Mobile is personal. Every person’s mobile is their own.
- The next mobile revolution is emerging markets: Africa, Middle East, parts of Asia, Latin America etc.
- The media have told us some truth about the third world countries, but not all the truth. It’s not all wars and hunger and suffering, there are many great things happening too.
- To solve the problem of poverty, don’t get the image of the poor into the minds of the rich, get the image of wealth into the minds of the poor.
- There are a lot of challenges when developing for emerging markets – coverage, power, device capabilities, lack of money or no way to spend it (no credit cards etc.).
Mobile is Not Different (@mattt)
- Mobile apps are internet apps, everything is connected to some API. A smartphone is not very useful without Internet connection.
- The more efficient we make something, the more it will be used, so we’ll end up using more of it in total (gas, coal, but also information).
- People looking at their phones will be a thing our generation is remembered for.
- IPv6 gives us an unimaginable amount of IP addresses, so in future everything will have an address, everything will be connected
- If your value is the content, make a responsive website; if your value is interaction, make an app. (But don’t treat it as a fight between good and evil, just use the right tool or even mix them if you need.)
Limits of Cross Platform Development (@robert_virkus)
- If there’s one thing that isn’t going away, it’s change.
- Challenges of cross platform development:
- different programming languages
- different device features
- in-app payments
- push notifications
- UI design paradigms
- fragmentation of OS versions
- Native apps:
- 1 team per platform, best integration but most effort required
- you need to find a way for the apps to share some algorithms, assets etc.
- some app stores may cause problems
- hard to optimize the UI for all the platforms at the same time (often the app is just optimized for one platform, usually iOS)
- uncanny valley – kind of looks like native but something is wrong in a disturbing way
- There’s no point creating an app that doesn’t do anything else than the website does, or an app that you aren’t planning to maintain.
- Choose HTML5 vs. native app depending on the specific app type, your audience, budget, developer experience, etc.
For a Future Friendly Web (@brad_frost)
- “Links don’t open apps” – websites are more connected, because you can link to any place.
- “You need to download our app to read this” – no I don’t, this is bullshit (Quora).
- It’s not scalable to make completely separate versions of a website for every device; make one version that adapts to the device (“adaptive design”).
- device diversity is not a bug, it’s an opportunity to reach more people
- content parity – give people what they want, give access to the same content and functionality regardless of the device; don’t prevent mobile users from accessing all of your content
- there’s an entire spectrum of resolutions, not just a few different ones; your site needs to work with every possible resolution, not just 320x480 and 768x1024
- ish. – a tool for testing multiple resolutions
- 74% of people will leave if your mobile site doesn’t load in 5 seconds
- 86% of sites send exactly the same content to all devices, including all JS and images, even if it’s not needed or not used (and it often weighs many MBs)
- performance is invisible, you don’t mock it up in Photoshop, you don’t schedule time for it while planning the project
- people hated the Facebook app because it was slow as shit
- Mobitest – tests your mobile site’s performance on a real mobile device
- design mobile-first, not mobile-last
- use tools like Modernizr to detect features
- if too much code is executed on the client, it might lead to slow performance if you overdo it
- there’s a difference between support and optimization, so don’t try to make it work perfectly everywhere, but don’t block people without giving them a chance
- people’s capacity for bullshit is rapidly diminishing
- people can circumvent any bullshit you put on their way
- get your content ready to go anywhere, because it’s going to go everywhere
- we don’t know what’s going to be under christmas trees in 2 years, but we still need to design for it
- the more backwards compatible you are, the more future compatible you are
- WTF Mobile Web – examples of mobile site epic fails
- Future Friendly Manifesto (futurefriend.ly)
- There are 6 bln mobile phones, but only 2 bln people have access to the web.
- When you’re finished changing, you’re finished.