Building an app or website is one of the biggest goals for new startups and solopreneurs, but many of us make key mistakes that stop us before we ever really get going. Here are nine common mistakes and how to avoid making them.
1. Going for the Big Bang
You've just had a great new idea for an app to rival Airbnb, Amazon and Facebook combined, so you take your life savings, pay an agency to build it and then discover people just don't get it.
Avoid wasting all your money on a white elephant by conducting small, low-cost experiments and properly analysing the results. Rather than building a fully-featured app that does everything for everyone, work out your quickest and cheapest Minimum Viable Product (MVP).
With an MVP, you can observe whether or not your audience appreciates your idea:
- Do they find it useful?
- Would they pay for it?
- Does it need to be tweaked in order to meet their needs?
As you build your knowledge further, you can be more and more confident that you are investing in a viable business.
2. Not delivering value
Successful apps, need a strong Value Proposition — a good value proposition helps your customers do the jobs that are important to them, generates unexpected gains and relieves pains. Value Proposition Design explains the importance of finding a fit between your customer profile and the features that your offering delivers.
Check that you are creating a desirable app with quick and easy experiments like customer interviews, search trend analysis or even spending a day in the life of a potential customer.
3. Not testing out with users
If you fail to test out your app or website with real users, you may find that they are unable to do the jobs they want to do and simply give up. This is particularly important if your app is going to be used by people with accessibility needs, but you may also be surprised that able bodied people also give up on things you thought were obvious.
Split testing two versions of a feature (A / B testing) can be incredibly useful for determining what one is most effective.
A number of tools are available for testing out the usability of websites or apps. These generally involve some form of task-completion script, so for example you might ask a customer of your adventure holiday platform to complete the following:
- Find a surfing holiday in Cornwall
- Book a holiday this summer for you and your family
Obviously as a usability tester, they won't want to actually spend money booking a holiday! It is important to set up a staging environment to ensure that your test subjects don't spend any real money and don't destroy your live app.
Depending on the platform you use, you can get a screen-recording of different user sessions. With some you may even get a video / speech recording to see reactions as participants progress through the tasks. Once the test is completed you can see your task completion rate and the average time it took people to complete each task.
Each time you deliver a new feature, get your customers to try it out — friends, family or colleagues are great for the early stages, but as you go on you'll probably want to source some other users. Thankfully, most of the user testing tools have a large group of users that can be selected based on things like demographics, browser and location.
Knowbility is a great source if you want to test out your app on people with disabilities using their Access Works database.
| Product | User testing platform? | User sourcing? |
|---|---|---|
| Usability Hub | ✅ | ✅ |
| Loop11 | ✅ | ✅ |
| Try My UI | ✅ | ✅ |
| Cint | ❌ | ✅ |
| Knowbility / Access Works | ❌ | ✅ |
Once your app is live, you can continue to observe how your customers interact with your web app with analytics tools such as Google Analytics or Hotjar (which records user sessions and provide heat map analysis).
4. Ignoring the evidence
It doesn't matter how beautiful your theory is, it doesn't matter how smart you are. If it doesn't agree with experiment, it's wrong.
Richard Feynman, Theoretical Physicist
If the experiments and tests you conducted don't confirm what you expected, it is very easy to just ignore them, telling yourself that your idea will work anyway. This is the wrong approach and you need to stop yourself going down that path.
Some experiments will generate weaker evidence than others, so in some cases you may want to follow up with another. The more experiments you do, the clearer the picture will be, so make sure that you respect that evidence and react accordingly.
By writing up a test card before you start each experiment you can make sure that you don't manipulate the results to fit your own beliefs. On the test card, set out your:
- Hypothesis: We believe that...
- Test: To verify that, we will...
- Metric: And measure...
- Criteria: We are right if...
To avoid confirmation bias, it can be useful to write a hypothesis that goes against your own opinions, so instead of "We believe that professionals will prefer using my amazing new daily planner app over MS Outlook Calendar", you might go with "We believe that professionals prefer using MS Outlook Calendar over my new daily planner app".
If the experiments you carry out disagree with your theory, you may need to alter your idea or possibly even throw it out altogether. Remember that it's better to fail early instead of investing time and money on something that no one is going to use.
5. Having no flexibility
When your evidence suggests that you should be taking a different approach, you need to be able to change quickly. To use an analogy from Eric Ries's The Lean StartUp, building a business should be like driving a car and not like launching a rocket. You should be able to make lots of small steering adjustments as you go — on some occasions you might even want to turn down a different road completely.
To enable an agile business, you need an agile app to go with it. Make sure that you can easily add new features and remove others without incurring ongoing fees. At Kingston Labs, we architect our apps using the JAMstack — a cutting edge approach that makes it easy to swap features in and out.
6. Scaling too early, or not preparing to scale at all
In a similar vein to going for the Big Bang, many startups try to scale their app too quickly. If you start your app with a huge database, numerous service subscriptions and an international network of web servers, you are going to incur massive costs right from the get go. Equally, businesses that do not prepare for scale, may be held back when their business starts taking off.
Start small and future-proof your app by using an architecture and hosting platform that can be scaled easily like Netlify or Amazon Web Services (AWS). Architectures that use serverless (lambda) functions are some of the most robust in terms of performance, cost and scalability — something that is offered by both of these platforms.
7. Reinventing the wheel
Developing custom-built platforms is great, but you should realise that you will need to support that platform for the lifetime of the app — something that could get quite costly as you start to scale up. If someone has already built it, you are probably better off using that!
With the rise of open-source programs and "Software as a Service" (SaaS), this is now easier than ever. To use an e-commerce store as an example, you are probably best off using a platform like Shopify or SnipCart. Paying a monthly subscription is far cheaper than paying to develop and support your own e-commerce system. Furthermore, they have probably already thought about all the little issues and features you might have otherwise needed to address along the way (like multi-currency, refunds, vouchers etc).
You may feel that your app or website needs to be more unique than anything that these services can offer, but with a bit of creativity and good engineering, a huge variety of ideas can be implemented.
For example, Shopify provide an API that lets developers build websites and apps that don't look like Shopify and still benefit from all of its key features including the store management dashboard. The power of this cannot be underestimated — it means that you get 24/7 support from Shopify (rather than depending solely on your developer) as well as regular security and feature updates without additional cost.
Of course there are some situations where you really do need a bespoke solution, but it is important to always consider the options available to you before committing to such a significant undertaking.
8. Not going to market quickly enough
If you are not embarrassed by the first version of your product, you’ve launched too late.
Reid Hoffman, Founder of LinkedIn
Your priority as a startup should be getting your product or service out. If you spend too long refining every angle you may find that a competitor beats you to your first customers or, even worse, you may have built something that your customers don't actually want.
Stay focussed on your key value proposition and build an app that delivers that — once your customers start using it, you may find that features you planned to develop are of little importance and a feature you didn't even think of becomes your top priority.
Choose a developer that can build you an app quickly and start testing it with your market. At Kingston Labs we have found that JAMstack has enabled incredibly rapid development.
9. If you build it, they will come
I always thought this quote was in reference to Noah building the Ark, but apparently it is actually a misquote from Field of Dreams with Kevin Costner!
The concept may work with mystical events but unfortunately it is very ineffective when dealing with an app or website launch. This has actually been one of the biggest issues for my own projects — you build what you think is a great app, but then find that no one is using it.
To avoid falling into this trap, build awareness of your project with a properly targeted marketing campaign. This will probably involve a social media presence, ads and Search Engine Optimisation (SEO). Ads on social platforms like Facebook allow you to target your audience extremely precisely so, for example, if you have a new baby product, you could target "mothers, aged 25 to 40 who are interested in baby food".
Drawing it all together
Hopefully you find these tips useful — just remember to test out your ideas early and only invest your time, money and effort when you have the evidence that people will engage with your product or service.
Do you have any tips for starting up a business, what's your experience been with website and app development? If you do, please leave a comment below.