A beach, some sky, a lovely message.

Backend as a service – is your app in the clouds?

As we peer at the detail of the latest World Cup stats from Google, for those not so keen on the beautiful game, it’s time to think about all of the many thousands of stewards running around behind the scenes keeping things together. Who does this for your app project? It’s time to peer into part six of our top ten tips for creating a mobile startup. Following up from our guide last time on how to find the perfect app developer / designer, we are now start to explore who is powering your app, who is behind it.

“Behind every great man, is a woman rolling her eyes…”  Jim Carry, quote from ‘Bruce Almighty’

As Facebook suffers it’s worst downtime of only fifteen minutes, it looks like the original face-stretching actor Jim C has also suffered the agony of trying to make sure his Backend as a Service (BaaS) keeps up with his frontend delivery. In essence every mobile app that is out there is a thin client to a much larger backend system, where data flows from app to database and back again over (hopefully) secured web services. A few years ago this would have required custom development as existing CMS tools could not provide the features developers required for apps. Now there are a range of cloud providers in the market, with the successful Parse snapped up by Facebook in 2013. What should you ask your provider?

 The power of the cloud for your app

Brooklyn Bridge
Behold the power available from the cloud that passes over us all
  1. Features, features, features – data storage is a given, are you going to provide me with any other features, such as push-notifications that are mobile specific?
  2. Your train is on platform 2, iOS only - do you have lovely SDKs or integration kits for every-mobile-OS-under-the-sun? Without Badu support, we are OVER!
  3. Price as transparent as water – can you see start for free and see the price levels clear as you scale up and down?
  4. Service updates and availability, better than air - what guarantees of uptime do you get for the service, has it been reliable in the past?
  5. Scorched earth policy, leaving and joining- is the sign-on process easy and how hard is it to get your data back out of the platform is you want to leave?
  6. The power of Yo, upscaling ability - what happens if your app experiences a surge in traffic up or down, is that a quick and easy process to upscale?
  7. Shop around for solutions - which are the best options out there to pick and choose from for your particular app?

Features, features, features

Some old cameras with film. Ahh, lovely. What features does your cloud-powered app need to have for your app to work? As you start to develop your app, whether in M.V.P. form, prototype or full blown app, you will soon want to store information in a database for users to get to later on. This may include:

  • Users creating an account, profile, logging in
  • Users storing data based on their app activity
  • Users making purchases, via in-app systems or externa
  • Geolocation for your app, with local lookups
  • Push notifications for status updates
  • File storage and access globally, including CDN and video

The cloud providers realised that there were different systems for each mobile eco system e.g. APNS for iOS and GCM for Android requiring developers to create two or more sets of code to support both. The solution? A cloud-platform based cross-platform solution to make integration easier. Before you choose your provider, list the key features you want to use now and what might be on your ‘wish list’ for later on to ensure they can support your plans for the first and any subsequent release you have there.

Your train is on platform2, iOS only

The advantage of using a cloud provider is that they support all of the platforms that you can think of. There are always going to be new surprises in the world of mobile apps, with programming langauges like Swift appearing, but in the most part the following platforms are still in use around the world currently so check your provider supports the following:

  • iOS for iPhone and iPad (as well as the new iWatch! Oh, spoiler alert there)
  • Android for all lovely smartphones and tablets from Samsung et al
  • Kindle SDK for the Fire-y phones and tablets from Jeff
  • Windows OS for all those shiny Microsoft Surface 3s shipping soon
  • JS support or native cross-platform OS platforms
  • WebOS for older Palm and other tablet devices
  • OSx for those two people still making Mac desktop apps

Price as transparent as water

The process of gradually increasing your userbase over time can also apply to your backend web servers, applications and databases. The theory behind cloud computing is that it is flexible and delivered to suit the needs of the customer, be that using teraflops of data to search for life on Mars – http://setiathome.ssl.berkeley.edu/ –  or to register and login a few thousand users on your app. What you want to be able to do is probably try a few services out for free to start off then ask for information about service pricing. This can be set on usage rather like traditional hosting or set based on number of users. If the former, expect the usual file storage, bandwidth in / out and speed capabilities to increase with cost levels. Check that the breakpoints can work with you expect your expected traffic to rise over time and that they rise gently so you can gradually increase your resources at a cost you can can afford. One final aspect to check is the cost of any support issues that you might want to raise. Most services will cover a few quick technical questions with some answers for free but if you have a more in-depth question or problem with the service, can you still submit that for free? Is there a paid version where you can pre-buy business level support?

Service updates and availability, better than air

A plane on the runway, waiting, not going anywhere.
We expect air to be there, how about your mobile backend service? On standby?

We don’t often wonder about how services like Google’s Gmail are constantly available, we just decry them when they go down for around an hour on one day. Your app users will expect a similar level of service when they start using your app, irrespective of whether it is a free or paid app, so making sure your backend service is always available is important. Ask your provider how they make sure their service is redundant and check what Service Level Agreement (S.L.A.) is provided by them, their guarantee of uptime. They may also provide analytics of how the service has performed in the past, although as with other services, past performance is no indication of future performance. Maintenance is always required to keep any service up to date so make sure you have subscribed to the right email list to check the alerts that providers will be issuing before they carry this out. Most will provide enough notice for you to alert your users, if they even notice, as many of the maintenance windows will be designed to be as small as possible hopefully and won’t affect your service to any major degree.

Scorched earth policy, leaving and joining

A lovely woodland scene. It's quiet, very quiet.
What do you leave behind if you leave your cloud service? Does your data survive?

After the big event, there is often a lot to clear up, unless you opt for a Burning Man style “take everything with you” policy. For future planning, it is always worth checking how easy it is to close your account and export all of your data. Do you have to give months of notice before leaving or can you just end at the month when you are all paid up? Are you able to extract all of your data in one fowl swoop in a download that you can then use easily in another platform, is the data machine readable at all? If not, is there a programmable way you can get at an export of your data via an API to save elsewhere? The other major issue is how to migrate your users over to the new service that you might be moving to, if this a bespoke backend service or even another hosted service. You may have to issue an app update to change the URLs you are using to point to your web services, if you aren’t able to simply update the DNS and map this over to your new service. In that case, it is worth sending out messages to your users well before time so that they are aware of the upgrade and also investigating whether to include code in your apps to automatically check for an update when they launch.

The power of Yo, upscaling ability

Recent apps that have launched and either flapped their way to the top, or yo-yo’ed their way up the charts, have seen interstellar growth in a very short period of time. Check that your provider is able to scale and what sort of timescale they would be looking at to do that. Can you resize your existing service to make it larger or do you have to migrate from the simple to the more advanced service? Do you have to factor in time for a server re-build or a DNS upgrade so that you service is unavailable for an amount of time? The advantage of using scalable services is that they can flex both up and down as well. There will always be upwards and downwards movements in your visitor numbers as promotions and other activity affect your traffic levels. You may be affected by the seasons or peaks in e-commerce about Christmas for example. Whatever the nature of your app, if you can reduce your service need at your downtimes, you will be able to save money over the course of a year. If you can predict when these up and downtimes may well occur, then some services also provide the ability to automatically scale up and down for you which is a definite bonus.

Shop around for solutions

There are a large number of providers of this sort of technology. Hopefully we have provided you with some of the main questions to ask each of them and review before taking the plunge with one of the solutions. Let’s run through the key ones to consider:

  1. Parse - the original service that started back in 2012 now owned by Facebook.
  2. Kinvey - a flexible service with some great developer resources on-board.
  3. Bassbox - a new entrant into the field, removing vendor lock-ins throughout.
  4. Kumulos - a service which focusses on strong customer support.
  5. Backendless - the instant backend service for mobile apps.
  6. Applicasa - a backend service targetted at gaming apps.
  7. Appcelerator - with over 600,000 developers in over 162 countries.
  8. Kony - enterprise level service with multi-channel cloud support.
  9. FeedHenry - backend services and app building tools provider.
  10. AppEngine - a flexible backend service from Google.

Next time…

Let us know how your search for a mobile backend as a service provider goes and send us your comments below. Next time? Come with us and learn how to, Tell your audience about your apponce you have developed it. If you have any questions or want us to cover a particular area in a future post, leave a comment below and we can follow up. Thanks y’all and see you next time! [Image credits via Unsplash ].

Want to create your own app?

mpu-wide If you have a mobile app project in mind and would like to get in touch about it, please fill in the form below, email us at info@bigorangesoftware.com or ring on 01737242329 – we look forward to hearing from you soon!

Some shoes, in a field.

App developers / designers – where you at?

As we count down the days to the next Apple WWDC and attempt to guess which company they are going to buy next, it’s high time to polish off the crystal ball and peer into part four of our top ten tips for creating a mobile startup. Following up from our guide last time on how to create your lean MVP for your mobile startup, we are now delicately tip toeing into the topic of how to choose your app developer or designer for your mobile startup project.

“As humans we have this innate need and desire to meet people,” he explains. “In the past, social networks were concerned with connecting you with distant members of your network of friends. Tinder is all about connecting you with new people. And we find that valuable…”  Justin Mateen, co-founder, Tinder

As Tinder starts a social revolution in the world of singletons in new and interesting ways, despite their best efforts many will struggle to use this in a business context. Trying to find the right app developer or designer for your project is always tricky due to the explosion of the whole app world since July 2008 and the launch of Apple’s first App store at that time. Forewarned is always forearmed we believe so before you step into the abyss that is developer-networking, here are some key questions to have up your sleeve. Let’s start with the key questions first and why you need to ask them:

You want to meet who…and why?

Cup, coffee, glass, desk.
Shall we chat apps over a coffee or two? Do you develop or design apps?
  1. Location, location, location – where are your new app workers and can you see, talk and chat with them easily, are they all definitely human?
  2. Together, apart and together again – should you get a separate designer and developer, bring them together or look for combined, superhuman skills?
  3. Cut me a slice and come on board – is it worth providing a revenue share agreement to an app designer or developer to bring them on board?
  4. A picture may hide a thousand words – you did create all that work in your portfolio and understand it completely throughout, don’t you? Really? REALLY?
  5. Daily commits, builds and bug checkers – what is this, why is this important and what does Joel have to say about it so that you can replicate his findings?
  6. Slot a goes into tab b - what is the design and development process and how should this be run to bring you the best and most appropriate ideas for your project?
  7. Show me the way to go home…and test – where testing meets UAT meets maintenance and support for the banishment of all problems
  8. The end is the start - where keeping pace with the ever changing app eco-system is helped or hindered by your choice of agency

Location, location, location

Desk with hipster equipment
Do you want to join the Crew? App market places allow both parties to meet up.

As the lovely Kirstie Allsop continues to push, when not releasing craft books of course, location is always an important factor when thinking about how to resource your mobile startup project. As is common now in software development for websites and e-commerce, there is an increasingly strong trend to source expert designers and developers from other countries outside of the UK due to a lower cost base. Sites like oDesk and Elance offer a wide range of support for projects and are even offering discounts for recruiting staff far and wide, with a range of advice and guidance about how to manage the process remotely without ever actually meeting the people that are working on your project. Even marketplaces such as Crew have started up to match projects to designers and developers.

In the last few years, we have also come across more and more hybrid entities in the mobile area. These tend to still use talent overseas, including Asia and now Russian countries, but have a head office and management function in the major cities around the UK. You get the opportunity to meet people at the company you have engaged for the mobile project, but you don’t get to actually see the people beavering away at the code for your project. The people you do meet may not have direct experience of creating smartphone or tablet apps and also have to relay all of your instructions and feedback across to them for any work to be carried out at length for you. This translation and distance in the relationship could well lead to complications where the fine details in projects is always important to create a great quality product.

In the end, the issue of whether distance is a factor for you or whether you would prefer to engage a UK-based digital agency to create your app is a personal one and is often influence by cost. If you are have a very clear idea of your app, have explored this in wireframes and sketches and have a smaller budget, then going offshore for your first version and any subsequent amends will save you money initially but may cost more in the long run with issues over quality and reliability to deliver. If you are at an earlier stage and want to explore your idea further and see how it fits into a wider campaign for your company or brand, it would probably be better to engage a UK agency who can meet, sit down with and scrawl on paper together with to get your ideas across and build a great app. In our experience, this is always more productive in the short and long term.

Together, apart and together again

Surely combining things is better than going individually, isn’t it? Two heads are better than one aren’t they? When it comes to app developers this ain’t necessarily so. The disciplines are fairly well divided in many digital agencies. The iOS developers don’t share much in common with the Android developers, neither of whom touch much Windows development at all. Designers are also rarely developers as well in our experience as they just employ different skill sets.

Engaging one agency to design and another to develop however, is increasingly common. Often the requirement to produce an app comes out of the more creative, design-lead marketing and promotion side of a business and so the two can be created separately. Ideally however, we would recommend keeping them in the same entity if at all possible. The heady goals of designers can sometimes not match the practicalities of the development team – that lovely feature to add a Facebook friend included in the full colour designs may not be supported by their API and if created separately are rarely checked together until it is too late and the deadline fast approaches. Chat to your prospective list of agencies, freelancers or other app designer and developers to see how they would approach this. Our recommendation is to make sure they can offer both sides of the app coin, with design and development in the same organisation.  The cross over between the disciplines during the early and even final phases will lead to a much better product in our experience.

Cut me a slice and come on board

Screenshot from iTunes
Is creating another Flappy Birds on a revenue share agreement the way to go?

A problem shared is a problem halved, as the old fairy tale goes…mind you, they also say don’t go into the woods at nighttime because of bears, or maybe that was something about picnics. Anyway, for those seeking to make the next billion from a clone of Flappy Birds, the temptation to share the risk for your new mobile startup project is strong. The rationale for entering into a revenue sharing arrangement with a designer or developer is that not only is the risk shared but so is the reward for both parties. The project will then fly out of the door as the whole team are motivated to create an exciting bug-free app release with amazing features and want to stay engaged with it as the sales and profits rise over time to a level where the helicopter becomes a realistic purchase.

While this may attract the smaller freelancer searching for their next project, for others and for many digital agencies this approach does have it downsides. For any app project to succeed in a competitive marketplace, every app requires dedicated promotion and marketing to survive and thrive over an extended period. In a revenue sharing situation the initial costs are lower which is great for the project owner, but unless they provide a significant marketing budget, the app designer and developer may struggle to recoup their costs over time from the ‘revenue’ element that is offered by the project owner so may not be as engaged as you would want them to be in your app project.

Even if you do find a freelancer or even organisation that does want to take your project on with this sort of agreement in mind, there are still some operational issues to consider. If they accept an agreement from you, they are probably doing it for other people as well. You will need to make sure your project is getting as high a priority as others and watch out for the situation which may arise if their other projects are making them more money and yours suffers as a result. If another agreement is giving them a greater share as well, the freelancer or agency may decide that it is more in their best interest to pursue that instead of your project.

A picture may hide a thousand words

Screenshot of the chupa website
The Chupa mobile site has apps-a-plenty but are they any good for your project?

With the rise in sites and exchanges that will allow people to buy and sell app source code such as Chupa, it can be hard to tell from a website portfolio whether the company or individual actually created all of their work themselves. Designers have always taken inspiration from other projects which is fine and developers have also always shared code samples – now even projects like WordPress have open sourced their app code to join other projects available so there is a wide range to choose from out there.

The question to be answered by any app developer or designer is whether they can present the pros and cons of their own apps that they have worked on. Are they able to explain the full stack of code that went into the project, which open source libraries were used, what features they included in any web-based admin panels and so on? If they had their time again, which elements would they maybe re-write or if they are working on a future version, what amends are coming into the project? Discuss this with your prospective app designer and developer to see how they view their previous projects. This will help you to understand what skills the app developer or designer has in understand the totality of the project and how they have been involved in it over time, which they can bring to your own app project.

Daily commits, builds and bug checkers

Screenshot from Joel's website
Joel on Software brings his 12 step check for software stardom – are they any use?

Primarily designed for job applications wishing to join digital agencies, Joel Spolsky listed twelve conditions that every software firm should have in place to create a great environment and produce wonderful projects, such as having a bug database. The list was created in 2000 so some of the techniques have been eclipsed by new practice, such as continuous integration replacing daily builds. On the whole however, the list does remain a useful starter when interviewing any app developer, to see how they cover key areas such as source control, bug tracking and testing in their projects. You need to feel confident that these areas are covered by their approach and that they have experience in these elements are they are core skills which should be applied to your app project.

Slot a goes into tab b

Post it notes at an agency
Is wireframing and userflow work in the development / design process?

If you have an app idea in mind, you may be anywhere on the spectrum of having nothing sketched out at all to having a full set of wireframes. When you arrive at your app designer or app developer, what is their typical software development process to run through to create your app project? From agile to waterfall and everything in-between, there are many buzzwords in this area and pros and cons from each. Most projects will run through the following elements however which you will need to make sure are included:

  1. Requirements gathering – what platforms / users are you targetting? how are you monetising your app project and promoting it?
  2. Design phase – do you have a set of brand guidelines / identity, how many rounds of amends are included here?
  3. Pilot testing – can you use tools to create an interactive prototype at this stage to test out the project and collect valuable feedback?
  4. Development phase – how is this managed, it is agile or phased in rounds? how can you deliver feedback on builds, how are these issued to you?
  5. UAT and pre-flight checks – how is the final testing run (outlined below in more detail) and who is creating the test scripts involved?
  6. Launch and test again – get ready for launch time!
  7. Revise and re-launch – check your user feedback, revise and re-launch to keep your app project at the top of the appstore charts

Show me the way to go home…and test

Over time the number of different Internet browsers has grown, making it hard for project owners to test their new web-based project in every single combination available. Virtual testing services sprang up to get around this problem with even Microsoft getting in on the party. The same issue now applies to mobile projects as the variety of Android smartphone and tablets grows daily and the two form factors for iPhone including the 4 and 5 shape joined with a variety in levels of iOS out there in the app ecosystem. Your app designer or developer should have access to test devices in their workplace or may choose to use one of the many mobile test studios available, even Google provide this at their Campus in London for those wishing to test out apps and mobile sites.

Testing thoroughly is important as many app designers or developers may require you to sign the work off as being complete and functioning before they release the final code to you. User Acceptance Testing or UAT is based on this process, where you have a list of the key functions of your app that you can run through in a series of test scripts to make sure you app is working correctly as you originally specified. Check with your app developer or designer to see how they include this in their process as after this is signed off, you will probably be entering into a support and maintenance period which is probably chargeable for any future work, depending on how they handle this in their organisation.

The end is always the start

The world of apps is constantly changing so for many seeking an app designer or developer, the thought at the forefront of anyone getting an app created is what happens after their app comes out. Make sure you check with your app agency about how they handle updates and amends to your app over time. As the Apple WDDC approaches for this year and iOS get another release, you need to check with your agency whether it can test your app in any developer preview release before these hit the store and your users start testing and letting you know of any problems.

New platforms and even app stores emerge over time as well, with Android moving into the TV space as well as Amazon and even Tesco rumoured to be creating their own smartphones. Chat to your future app agency to see how they keep in touch with these changes in the landscape and will communicate these to you so you can keep your app project up to date and reaching as many customers as possible.

Next time…

Let us know how your search for an app developer or designer goes and send us your comments below. Next time? Come with us and learn how to, Match your app to the power of the cloud. If you have any questions or want us to cover a particular area in a future post, leave a comment below and we can follow up. Thanks y’all and see you next time! [Image credits via Unsplash ].

Want to create your own app?

mpu-wide

If you have a mobile app project in mind and would like to get in touch about it, please fill in the form below, email us at info@bigorangesoftware.com or ring on 01737242329 – we look forward to hearing from you soon!

 

 

Picture of a shed

Lean? Mobile? MVP? Me?

Surviving the hordes at Barcelona’s annual Mobile World Congress stampeding the Samsung stand for Galaxy S5 news and sidestepping Neil Young’s PONO audio launch at SXSW, we bring you the shiny launch of part four of our top ten tips for creating a mobile startup. Following on from our guide to mobile analytics and why you should bother, we move swiftly on to the latest buzzword in startups, becoming ‘lean’ and creating your mobile MVP. What does that mean for me I hear you wonder?

“Hide not your talents, they for use were made,
What’s a sundial in the shade?”
― Benjamin Franklin

Stay lean, MVP, wherever you can

A mountain
Everyone has to be lean and green when it comes to mobile startup MVP projects.

The concept of being lean seems to apply to most things in life and has now hit the startup world. Is your project lean? Well, consult the following lean principles:

  1. Entrepreneurs Are Everywhere – you don’t have to work in a garage to be in a startup.
  2. Entrepreneurship Is Management -  A startup is an institution, not just a product, so it requires management, a new kind of management specifically geared to its context.
  3. Validated Learning - Startups exist not to make stuff, make money, or serve customers. They exist to learn how to build a sustainable business. This learning can be validated scientifically, by running experiments that allow us to test each element of our vision.
  4. Innovation Accounting – To improve entrepreneurial outcomes, and to hold entrepreneurs accountable, we need to focus on the boring stuff: how to measure progress, how to setup milestones, how to prioritize work. This requires a new kind of accounting, specific to startups.
  5. Build-Measure-Learn – The fundamental activity of a startup is to turn ideas into products, measure how customers respond, and then learn whether to pivot or persevere. All successful startup processes should be geared to accelerate that feedback loop.

For many mobile startups, these principles apply to two key areas. Firstly, the people you need to bring on board for your project to get things up and running, during your first year and secondly, the product you need to be able to build. Taking the people side first of all, would would be your M.V.T. or Minimum Valuable Team…three you say? Oh, you mean…

A Hacker, Hipster and a Hustler…

Man skating across concrete.
To create a mobile MVP, your project needs a hacker, hipster and a hustler.

These three areas break down into the three key focus areas for any new mobile startup. The hipster deals with design and user experience, how the app is going to look compared to others on the market, current trends and the brand. They also make sure the app feels good in the hand, that it meets the overall business purpose and looks great (with or without hipster facial hair of course). Expect a strong view on everything from the Windows Metro interface to the one year iOS7 birthday and the colour of their shoes.

The hacker, clothed in black and spare pizza, eats, sleeps (rarely) and breathes developing, working on all of your coding as well as creating the backend systems that allow you to control your mobile app and find out about your users. They will keep up to date in trends and convince you to use *insert-name-of-new-database-software-here* without really being able to tell you why. They won’t like sitting next to hipsters.

The hustler is a more precise animal, often spending large amounts of time out on the road, looking for new business opportunities and partners for your app project. When they are back, they will be busily updating social media, amending the latest marketing plan to work out how to tell which demographic about the amazing new app you are working on. They will be constantly pushing the hacker and hipster to move faster towards the “we are selling to Facebook for 1 zillion dollars” moment. This may not come. Possibly.

Next time, we will focus on where you can get these skills if you only have one or two of these areas covered, which you can outsource and which you should try to grow from within, if that is achievable on your project.

I have people what do I build? An M.V.P.

A computer and desk, with a notepad.
The tools of the job? Laptop, notebook, pen. Done! What to build though?

Once you have your team in place, creating a first version as soon as you can is imperative. We explored earlier some new techniques for creating interactive app prototypes so hopefully, Blue Peter style, you have done that already before arriving here.  The reason why people focus on their Minimum Valuable Product is the realisation that you won’t capture everything in version 1, but you need to have version one out there in the world so that you can listen to your users and hear how things go.

  • To that end, what are the key features for your app?
  • In your elevator pitch of thirty seconds, what do you list and talk about?
  • If you had to get that down to ten words, or five, what would they be?
  • Which features are essential for your app, which desirable?
  • What features would you like to have after one year, two, five?

Once you have that list of features boiled down to the key ones, you are ready to build your M.V.P. without too much worry. The team at 37 Signals often mention in their Getting Real book focusing on lean software development about how handle request for features. The plain and simple truth is that your users will contact you repeatedly about the features they really want and rarely about ones that don’t matter, allowing you to jump and plan accordingly, but only once you have your M.V.P. out there, for review.

Release early, release often

Boats leaving a harbour
Why should you release early and often for your mobile MVP? Read on…

Once you have your shiny, new M.V.P. , lucky you, how and when should you release it? The mantra of startups has always been R.E.R.O. or relase early and release often, from the wise words of Eric Graham. Traditionally cited as one of the ‘hardest lessons for startups to learn‘, this process of pushing out releases more often that either the hacker or hipster may be comfortable with does one valuable thing. It shortens the feedback loop, getting you valuable information back from the people that really matter, your users. As Matt from WordPress even vouches, usage is definitely the oxygen of ideas, the trick now is how to capture those and when to record them.

Life in perpetual beta

So, because you are always testing and releasing your app project, does that mean you are in perpetual beta mode or should be? Gmail is the industry standard joke on the subject as it appeared to be in Beta for about five years or more. It did however release feature after feature, platform after platform and continually develop during this period. Heck, after they eventually rolled out, they even allow you to turn the Beta logo back on! Love ‘em. espite the recent film about this, being in this particular phase of development doesn’t really matter as long as you are constantly developing and listening to user feedback while you do so.

What do you mean, does it come in white?

Channeling user feedback is always difficult but necessary to provide some direction as to what phases should come next in your software development. We covered how to create a community for your app project and why previously, as well as the tools such as Testflight and HockeyApp to distribute and monitor your releases. Since then, Testflight has been purchased by Apple so watch this space for future development. Android apps can still use the free and great beta release service that runs from the Google Developer console. Even Mark’s using it for his Facebook App on Android at the moment so do check that out.

Do I have a Minimum Loveable Product?

At the end of the day, you need to make sure you still love your product. Lawrence from Spook Studio talks about his Minimum Loveable Product, Hiut Denim just try Do One Thing Well and even Google try to do One Thing Really, Really, Well. By keeping your app startup lean you will create an amazing M.V.P. and take over the world. Well, one at a time anyway. Let us know how that goes and send us your comments below. Next time? Come with us and learn how to, Choose your design and development partner wisely. If you have any questions or want us to cover a particular area in a future post, leave a comment below and we can follow up. Thanks y’all and see you next time! [Image credit "Bike outside house" - Florian Klauer via Unsplash ]

A person standing on a cliff, looking out to sea, pensively.

What is app analytics and why should I bother?

Stepping through floods, high water and the remains of the turkey, we emerge bleary eyed into the new year to open the wrapped present marked part three of our top ten tips for creating a mobile startup. Following from our guide to creating a community around your app, we arrive now at the sticky issue of all things number shaped, analytics style.

“How can you get 1 million paying customers? Easy! Get 1 billion users!” Phil Libin, CEO, Evernote

Digital marketing is full of analytics in a way that traditional media still struggles to compete with. Hits, plays, views and downloads are common phrases in the app world, joined recently by the key metric for apps, the Average Return Per User or Per Paying User. To work out whether your app is a success or not, you need to break down the key questions that your data needs to answer, then look to see what solutions are in place to help you measure that. Let’s start with the key questions first and why you need to ask them:

You want to know what…and why?

  1. Which platform and device are people using to access my project? 
    During 2013 we saw increasing Android growth to the point where even Android tablets overtook iPads. Typically a mobile startup will be tightly focused on their budget from day one so choosing the right platform and device is key. Our advice here is always match this to the user your startup is targeting and check the most up to date stats of app device ownership and usage you can before you start.
  2. What level of software OS are they using?
    As soon as Apple released the wonder-i-Cloud with iOS5 then the Facebook-friendly-iOS6 which left behind the first generation iPads, the issue of what level of software version your app users will be running came to the fore. Apple now display the level of iOS usage from store analytics, with iOS7 running at over 75%, but if your app is targetting older devices, you need to check whether that unique feature is supported for those devices. Android fans have similar issues with over 75% of their users running over version 4, but arguably better support for deprecated features.
  3. Which country are your users coming from?
    Localisation is now a key requirement for any mobile startup creating an app that instantly has access to a global marketplace. Both Google Play and Apple iTunes allow a rollout across the globe with the ability to create both a store listing and app content in other languages. The key issue for an app owner is to find out which countries their users are coming from and then assess which languages need to be employed. Companies have now sprung up to offer this service, with Google even including this in their app store listings as part of their analytics.
  4. How many new users are you adding over time?
    The golden numbers of any project are new users, being added monthly, so this is a key metric to ensure you are recording. You can then demonstrate user growth as a result of marketing and other promotional efforts you are making so it is worth recording which dates these are made on so you can correlate between the two at a later date.
  5. How many active users do you have each month?
    Following on from the number of new users you are adding each month, is this key metric, whether they are actually active or not over a set period of time. This will differ wildly depending on the nature of your app project. A photo sharing app would expect usage every single week from their analytics, whereas a pet vaccination reminder system might have a monthly or even annual usage base.
  6. How long are people using your app for each month?
    App duration is increasingly important to record, to gauge how sticky or attractive your app is for your intended audience. It will also help you attract in advertising or sponsorship, depending on the business model you have chosen for your app, so you can convince others how many eyes are regularly visiting your app to use your service and may see that brand or organisation willing to get involved.
  7. Are there any peak times of day or week for use?
    The longer that your mobile startup is trading, knowing your peak time of day or night that your users are engaging with you becomes more important. As you start to grow from one country across the globe, rolling out an update or even a price difference requires changes to be made over several days before everyone will see it. It will also allow you to schedule your maintenance windows to target the times when users aren’t busy. With scalable cloud providers powering many apps rolling out autoscaling, even your servers can change to match times of peak traffic, so making sure your analytics can capture this is important.
  8. How buggy is your app for your users?
    For those app owners without a technical knowledge, it can be hard to assess how solid the codebase is that your project is founded on. The speed of agile development often leads to technical debts where code is created quickly to suit a purpose and never re-factored on a stronger footing. As a metric, if you record the number of support tickets or incidents that you get reported then divide this by the number of active users you have, voila, one instant buggy metric! This also helps you project how much time and resource you may need to dedicate to supporting your users as that particular metric grows as well.
  9. What is the average return per user?
    For those paid apps or those with in-app purchases that seem raise the money, the key metric is the A.R.P.U. or Average Return Per User, extended to Per Paid User in some cases is the key element. If you have a user-based method of supporting your startup, then this is simply a product of the revenue you take in each month versus the active user base you have during that period. If you are supporting the app via sponsorship or advertising, then this may still be an interesting metric for you, but altered slightly, to see your average earnings per user to then evaluate whether a particular marketing effort is worth pursuing to add each new user, based on the advertising or sponsorship value they may have.
  10. What is the overall satisfaction level with your app users?
    One for the office wall really, to keep your team focussed on making the best product in the world. Retailers and product manufacturers employ third parties or their own vast teams to carry out user surveys to gauge their customer satisfaction levels. App owners are already one step ahead here as you have access to appstore ratings for each version you publish. These are famously  inaccurate so a quick user survey may be a better source for this, to help show the human side of your app users.

Okay, before you start to panic and wonder how on earth you are going to gather all that information, just stop. No, not Hammertime, or time to Collaborate or Listen. Let me run through some of the tools you can use to gather this information so that you can create your own infographic of app use, and ride the analytics wave to success.

 How on earth do I find all that out?

1. In-app analytics

Launch an app onto any app store without analytics is madness. Most if not all of these services are free, or have a free account level including ones such as Google Universal Analytics, Flurry, Localytics or BugSense. Each will provide you with user data (new and recurring users) as well as devices (type and operating system) as well as custom goal and event tracking (users clicking on a particular button or using a feature).

Discovery analytics are also becoming increasingly important to add into your app. How did people come to download your app, was it from a website or social media? There are now services from both Tapstream and Parse to allow you to collect this from within your app by including additional analytics services.

2. Appstore analytics

Once your app(s) are published, analytics data can be gathered from the appstores directly about your userbase and sales. This will help you calculate the sales and values per user metrics listed above, updated automatically to include new data. Services such as Distimo again offer a free account for you to use, but they do aggregate data and sell this commercially, in their annual reports and more. Alternatives include Mopapp, AppAnnie and AppFigures.

3. Your users

We covered earlier how to recruit a community around your app and why this is important. One of the great sources of data are the testing services we covered such as TestFlight and HockeyApp. If you open up a beta phase and invite testers to apply, the services automatically log great analytics including user data to help you answer questions above about which devices, platforms, software levels and countries are being used in your mobile startup. If not, tools such as SurveyMonkey offer a free service which is optimized for mobile devices so you can create a quick, simple survey to gather user opinion then include this in the app and highlight it to your users.

4. Other apps or Benchmarking

Most mobile startups are often aimed at doing something better in a busy technological sector rather than being a true innovator. Whichever side of the divide your startup falls under, you can start to benchmark your app(s) against others in many different ways using analytics. Services like MobileDevHQ allow you to add your app then compare analytics with other apps, your competitors, in your particular category to flag up trends and changes in data.  Many offer app store optimisation services as a result, there are even enough firms out there competing for your apps, that there is a top ten list of providers, all with some level of free and paid accounts.

Okay, I get the picture, but how many users is enough?

Many mobile startups struggle with this particular question and obviously, it varies depending on the overall goals of your organisation and your company size as well as your overall view on analytics. One interesting answer to this recently was posed by the startup guitar-based network Strumm. For them, they felt that having over 1,000 meant it was a ‘thing’ for them and proved they had enough support from their community to take it further. How many and of what do you think you need for your app-based mobile startup? Let us know and send us your comments below. Next week? Come with us and learn how to, Choose your design and development partner wisely. If you have any questions or want us to cover a particular area in a future post, leave a comment below and we can follow up. Thanks y’all and see you next time! [Image credit "Bike outside house" - Florian Klauer via Unsplash ]

Engaging the community for your mobile startup project

Creating a community around your app

We continue in our quest to reach the final goal of our top ten tips for creating a mobile startup, arriving at part two. Following on from wonderful top ten tips for your mobile startup,  this week we tackle the community. What is it and why is it important?

“And, you know, there is no such thing as society. There are individual men and women, and there are families”.  Margaret Thatcher, 1987

Roll forward a few years and unless you have been hiding in quite a remote cave, you will have seen the world of crowd funding arrive and develop for a wide range of businesses, creating a new set of economic families. Kickstarter now has everything from mobile apps to Christmas glow in the dark t-shirts. Luckily not all are funded, but the best are. Involving the community of users that might take up your app is a great way of starting the development of your mobile app. If you still need convincing, let me run through some of the key questions mobile startups tend to face and see whether the community can help with any of that. Here goes:

  • What devices and which versions do you need to support? – the community can help.
  • Which features should your mobile app have? Ask the community.
  • Who is going to download your app from day one? The community. If it is there.
  • How will you know what bugs and issues need fixing in your app? Ask the community.
  • Why will you be able to charge users for your app or service? Because you asked the community.
  • How do we answer the quest for world peace? Ask the…hmn, maybe just get Kevin Bacon on that one…he seems to be able to connect everyone these days. Or that lovely Van Damme Man, he seems to build bridges everywhere.
A concert, with lights, fans and an empty stage. The community awaits.
The community is waiting for your mobile startup to launch. Is it on yet?

Hopefully are all now all down with the community. They have turned up. The stage is lit. What to do, what to do…practically, there are three useful steps to take now to start engaging with them over the mosh pit, with / without leather trousers:

  1. Online community engagement tools
  2. Community engagement website
  3. Test community recruitment

1. Online community engagement tools

There are a variety of great customer / community engagement tools around product support. Uservoice, GetSatisfaction and other similar services such as UseResponse and UserEcho aim at providing an independent service to owners of product and services, to support their user community. Suggestions, amends and ideas for developments and improvements from users are all allowed to have their voice here with one main difference. The clever part is driving users to vote on their favourite amends to allow the business owner to review the current app then start to plan for the features to roll out over time, to satisfy the needs of the community.

A screenshot from the National Trust UserVoice page, showing their community of users.
Want to involve your community in your mobile startup? The National Trust do.

Most services included here have APIs available too so they can be embedded within the actual apps themselves e.g. the UserVoice iOS SDK. This means that app users can engage with you as a business owner without having to leave the app and more importantly without leaving negative comments on the app store and rating you badly. By providing a community and support service easily to the app user, as they are using the app, the barrier to interacting with you is set very low and engagement is much higher.

2. Community engagement website

Once again, there are again a variety of different services you can choose from including the hosted Launchrock and KickoffLabs or even a WordPress powered solution via Launcheffect App . All of them have a common goal, which is to provide one central point for users interested in your app project to register their interest before you launch. Typically this fires off their details into an email newsletter, for you to send out details at your will, but also this tends to include social media links as well so the users can choose how to keep themselves updated. Design-wise the idea tends to be less = more, with the intrigue being built up by large background images and esoteric marketing strap lines to tease the user. Often this is because Jorge the ROCKSTAR DESIGNER still hasn’t work out which shade of teal your app icon needs to be and wants it to be featured on http://iosicongallery.com/ as soon as possible. While all that is being decided, best probably to keep the design and layout simple as you can always update this over time to match the app as it develops.

A screenshot from the launchrock website, showing two test devices and engaging their community of users.
Want to launch your app the rock way? Check out the launchrock service, a hosted service.

I would advise here, if you have diligently followed my previous step and established a presence on any of the community tools I mentioned, to simply link out to that here instead of embedding it. Why? Well, users can be unpredictable and can create content at will at a speed that is terrifying and strange in equal measures. As a result, you don’t want to put off new app users, potential funders, advertisers and beyond by showing off the extremes of your user community, but you still want to have them involved. Keeping this on a separate site as well allows community members to feel a sense of ownership over the content and means the feedback or interaction should be more informal and truthful. This might be painful but it is also beneficial in the long run to provide you with greater insight into what your users are thinking about your app at that time. Starting this process early and getting your URL up and running as well provides additional SEO benefits as well. If you have a working title for your app project that there could well be competition for, starting the holding site allows you to start capturing some of that traffic as well as allowing you to collect analytics about site visitors. If you do have a name or brand decided, even trademarked no less, you can add your app name into the store before you have even cut a line of code. As long as you follow up with the real app in less than 180 days in Apple Land or forever if you are in Google Land;  you can leave that there as a placeholder to make sure nobody else produces an app with that name.

3. Test community recruitment

The third and final strand I am suggesting is around establishing app testers from within your user community. Hosted services were set up a few years ago to use the Apple iOS features for O.T.A distribution of apps to users, the leaders being TestFlight and HockeyApp, whereas bug trackers such as BugSense, Crittercism and Applause also handle some of these elements. The main aim of both is to allow you to test your app with your community during a beta or pre-live release. In both systems, you set a recruitment link here and here, on both systems respectively. You can then send these links out to your community however you wish, for them to sign up to. When they sign up, you will then get a copy of their UDID to pop into your provisioning profile, but more importantly information about their level of iOS and also their device. You can then login and see what kind of split you have across both of those elements, to check whether you are supporting all of them or whether you actually need to! Once you issue the app to those users, their feedback and more importantly app crash logs, will be invaluable to help you improve the performance and quality of your app before it goes live in the iTunes store.

A screenshot from the TestFlight website, a useful community engagement tool.
Want to recruit your app testers from the community? Come fly on TestFlight.

These services include Android as well as iOS platforms, but luckly Android only apps could well use the new Google Developer service for beta testing, rolled out first at the Google I/O Conference in 2013. This provides you with a range of advanced tools to do rollouts for testing, including:

  • Community engagement – setting up a testing group via Google Plus or Google Group
  • Alpha and beta stages – providing flexibility in rollouts
  • Staged rollouts – giving apps to percentages of your group over time
  • All running through Google Play – for ease of access and use
  • Crash reporting as standard – all included for free

There is also a limited automated testing service in open beta from Amazon for its Android / HTML Appstore - currently it doesn’t involve community engagement but watch this space. I am sure as they have figured out those drones, that will be the next amend on the cards. Subscribe to our blog for updates as we unpick these top ten ideas and give advice and guidance about how to tackle these in your startup. Next week? Come with us and learn how to, Work out how to measure success before you start. If you have any questions or want us to cover a particular area in a future post, leave a comment below and we can follow up. Thanks y’all and see you next time! [Image credit "Bike outside house" - Florian Klauer via Unsplash ]

Setting up a mobile startup

Creating your own mobile startup – our top ten tips

As the First Lady of Country once said, working nine to five is a hard way to make a living. Heck, even the Daily Telegraph agree that this is a thing of the past but what has this been replaced with? The five to nine of course, creating your own start up business in your bedroom to follow the Zuckerburg dream and route to success.

“I think the biggest mistake that we made, as a  company, is betting too much on HTML5 as opposed to native… because it just wasn’t there - Zuckerburg”

Well, nearly always success. Here at Big Orange Software, where we develop lovely, shiny mobile things called iPhone apps and Android apps, we have worked for a range of startups over the last couple of years and have seen a real growth in this area. If you have a great app idea and want to take this further, let me run through my top ten tips or everything you wanted to know about setting up a mobile startup but were afraid to ask. Here goes:

  1. Follow the DIY approach and build your own prototypeI will run through a range of tools here from paper and pencil to digital wireframing software to show how you can create your own free, clickable prototype to share and wow your friends with.
  2. Create a community around your product early on - learning lessons from the crowd-funding business model, I will show you how to gather your community together before you release your first app, as they are your future customers and funders.
  3. Work out how to measure success before you start  - analytics drive modern digital businesses these days so deciding what statistics you will need to collect for your infographics to include in your future pitches at the start is key. I will run through some examples here and discuss how to measure these.
  4. Adopt a lean startup approach in everything you do – the release early and often mantra now extends to businesses with a lean startup approach. I will explain how you need to question each feature request and use your customers and their activity to drive what you need to include in your mobile startup.
  5. Choose your design and development partner wisely - from incubators, to offshore suppliers, freelancers to large digital agencies, how do you choose the right partner for the design and development of your app? I run through some check lists of key questions to ask before you commit your money to anyone.
  6. Match your app to the power of the cloudmobile apps are increasingly integrated into websites, across other devices and social networks or push notification services. I explore hosted cloud-based solutions for data storage and messages and look at some pros and cons for hosted versus self-build to see what is best for your mobile startup.
  7. Tell your audience about your app - if the hard work of producing your app has been done, how do you tell your audience about it? There are several low-cost options for marketing your app to mobile users as well as other opportunities to maximise the PR you can get. I outline some good practice in this area with some case studies.
  8. Support your product and your customers – users downloading apps for free or paid expect a service and support mechanism, like any other product purchase, and will soon rate your app badly if you don’t respond. I will describe a couple of hosted solutions and amends to your app that will increase the support you are giving.
  9. Be clear about your funding pathfor anyone signing up and joining you on the journey for your mobile app startup, be clear in all of your marketing materials and collateral about how you are going to be seeking funding for the platform. I will describe several options you have here and how to represent these online.
  10. Show your passion for everything – you are probably starting this new business because you have a passion for it and believe in the goals you have set. Don’t hide those under a bushel as I explain here how in the mobile startup world, those are one of your greatest assets and can help you spread socially to reach customers.

I will tackle these in individual posts over the coming days, weeks and months and expand on them in a bit more detail to help you to understand what steps you need to take with your mobile startup. Let’s tackle number one first:

1. Follow the DIY approach:
build your own prototype

If you have an idea about an app, it can often be something to solve a problem that you see with the world around you, a handy way to save your children’s artwork from the fridge before it meets the bin for example. Whatever the concept is, you can start by building it on a paper with a pen or pencil yourself.

Artkive - children's art app
Check out the app to collect childrens’ art – a startup mobile app.

Start by creating a sitemap first of all, a map of all the pages you might need. Where do users start? Is there a splash page first, then a login box and registration form? Is there an error message if that fails? Where is the home screen? Lay all of these out first of all and see how things are looking. Want some examples? Google is your friend as usual, see some of these app sitemaps, or an online sitemap generator to do the trick.

Once you have the overall map of app pages, you can start to sketch out how these might look. For analog fans, you can use an iPhone stencil or an iPhone stamp to get sketching. For digital fans, you can try either some of the great iPhone Photoshop templates available, or even kits for PowerPoint or Keynote to get you up to speed. Oh and for you hardcore Visio fans, there are even kits for Visio mockups as well.

Once you have your screens built, I would highly recommend building your app in the new, wonderful *no referrer fees paid to us here* Paper On Prototype service, available on iPhone as well as Android. It has some lovely features, namely the ability to share online with others for feedback. Check out the sample  POP app mockup to see what I mean.

Paper On Prototype - the app to make apps with
This is a great solution to make app prototypes using an app yourself, POP app.

This is a great way to create a clickable prototype that you can then share with others to get feedback. The only cost to you is your own time and maybe the odd pencil, but it means you can do some early user testing with friends, family or any other interested parties before you start cutting any code at all. You can revise this as well, taking out pages you don’t like and adding in new ones to get the app right.

Subscribe to our blog for updates as we unpick these top ten ideas and give advice and guidance about how to tackle these in your startup. Next week? Come with us and learn how to, “Create a community around your product”. If you have any questions or want us to cover a particular area in a future post, leave a comment below and we can follow up. Thanks y’all and see you next time!

[Image credit "Boat on Water" - Anton Sulsky  via Unsplash ]