App development, getting started. Do you really need one?

App development, getting started. Do you really need one?

I’m often asked: “can you build me an app?” People tend to ask this question with an idea for an app in mind, that in a dream world would allow them to quit their day job and become CEO of a startup company. The standard answer is, of course, yes we can build you an app. What people tend not to recognise is that the development is rarely the complicated part of launching an app. What’s difficult about launching an app and making it a success is the planning, user research, financing, marketing, launch, support and scaling. All these things need to be taken into account before it becomes a market-beating product.

Do you really want an app?

First of all when people say I want an app, what they tend to mean is I want a website. An app is just a fancy way of saying that.

It’s worth noting that this article assumes you’re intending to build an app that provides a service of some sort, rather than a game / augmented reality platform / something equally wild.

Let’s look at the downsides of developing an app from scratch.

First of all, app development is expensive. It tends to be a specialist skill, it’s often a more costly skill to hire for than web development and the whole process is significantly more time intensive than making a website. So from the off, you’re looking at having to dig deeper into your pockets.

Secondly, apps often have to be developed for the environments you intend to release them in. There are some ways around this (Ionic, Phonegap etc), but depending on what you have in mind, native development may be the best route. Even when using Phonegap, there can be significant work involved in ensuring the builds work consistently on both iOS and Android. Submission to the app stores themselves can be a time-consuming process and can lead to last minute crises in the go live process.

Even the building of an app within a single ecosystem can cause significant issues, as due to the maturity of the marketplace there is a huge variety of screen sizes, operating system versions, chipset etc., although this issue is slightly simpler in iOS than in Android where multiple manufacturers also come into play.

I don’t need a website

So you might be thinking, I definitely just want an app, not a website. Well, wrong again. App store deployment requires that you have a website to list terms and conditions and as a source of support information. Secondly, you’ll need a place to help you market your app, sell the benefits and explain what it does – just listing on the app store won’t lead to sales without some marketing in this will happen in the wider world.

You’ll also need to consider users who don’t own a smartphone or use a platform that’s not worth the investment to develop for (e.g. Windows phone) – do you really want to ignore these potential customers? Even the users who do have smartphones will occasionally want to access the service from a desktop computer, and an app won’t do them much good there either.

Finally, an app is now very rarely a standalone tool. Almost all of the time, an app needs to talk to something. Data needs to be stored centrally in the cloud, often with some level of synchronisation, user accounts need to be managed, information needs to be exchanged, the service needs to talk to other systems like APIs or social media. All of this means that the underlying architecture that supports the app is really a web app.

Ok, this is looking pricey

So we’ve already established that you’ll definitely need a site, as well as the app you originally wanted. But could you do it with just a site? The answer is yes.

Just building the web app, responsively and set up to be stored to the home screens of devices is often more than enough to test whether your idea for an app works, at a third of the price.

There are a few things that a web app will not do that a native app will do, including push notifications and talking to some of the device’s hardware e.g. accelerometer, gyro etc, although there are some ways around these, for example, sending SMS messages instead of push notifications.

Build the minimum you have to

Building a Minimum Viable Product (MVP) often just requires a web application as this will deliver nearly all of the functionality you’d expect to see in your final consumer product, develops the underlying web service required to run the application, can be tested by users on a variety of devices, can use web libraries for rapid front-end development, can use Javascript to power the synchronisation with the web app and 3rd party services and can be developed cheaper and quicker than an app would. There are no issues with deployment and no app stores to run it through. This allows you to be more agile and allow you to test the idea of your app in front of an audience – possibly getting you to the stage where it could even generate revenue.

Most would be app founders will require external funding to get their idea to fruition (as app development can cost in the hundreds of thousands of £/$), but in the UK investors tend to like to see some level of revenue being generated before investing in a  new product. Getting your MVP to this level can be critical.

A word of warning

One thing to note about this approach is that to a certain extent an MVP should be quick and dirty. It’s the only effective way of using the budget to get your product out there. If you spend months and half the budget on design, then you’re not using the budget in its most effective form. This comes as something of an anathema to some (more perfectionist types) but it’s crucial to be harsh on yourself in order to go to market as soon as possible. Move fast. Break things.