Get the best value and outcomes from your QA testing programme
Testing to find issues that might impact your users is a vital element of your software development process. Without it, you take a risk that could jeopardise the success of your software upon launch.
But in a dynamic digital world where development holds the key to exciting customer experiences and commercial innovation, focusing on the negatives is an unpopular pastime.
Bug-finding shouldn’t be regarded as a hurdle to jump to finally get your software out. It’s essential to put your software through its paces by testing out all kinds of niche and inconvenient user and technology scenarios, to make sure it stands up to real-world usage. As well the negative act of picking up on potentially damaging issues, testing provides you with risk model data for future software testing. That’s a very positive input that provides ongoing confidence in the speed and rigour of your future digital testing.
In this blog, we’ll look at some of the reasons why bug finding has negative connotations and recommend some positive approaches to make sure you get maximum value and results from your digital testing.
Why testing gets a dreary reputation
Is software testing is a classic grudge purchase? No-one gets up in the morning feeling overjoyed at the prospect of putting their software through a bug-finding process.
It would be great if you could guarantee that all your dev and digital projects would be issue-free and we could all experience only the positive excitement of launching a flawless application or feature upgrade. But that’s not real life.
Where testing takes place at the end of a release or sprint, it’s tempting to rush it or not conduct the depth or breadth required because the bug–finding is spoiling everyone’s fun by holding up the release or launch.
Test earlier in the SDLC
Testing should be built into your ongoing QA processes. The shift-left approach puts QA and testing into the mix as early as possible in the SDLC. That should reduce the time required for testing immediately before launch and allow you to identify many issues earlier on, so they aren’t replicated or baked into the final version.
Exploratory and regression testing should both be business as usual during your software development lifecycle. Development teams need clear, actionable and proportionate feedback from intelligent testing teams. Testing should be designed to help them improve the quality of their digital product development and enhance their reputation. That’s makes for a win-win!
The £££ business case for doing testing right
Across all industries, in the US alone, software bugs cost an estimated $2 trillion per year. That’s a big number.
If you’re in the online consumer retail business, the cost of software bugs will hit you particularly hard, because of the size of the market and the potential cybersecurity risks of errors in transactional apps and websites. Just one bug can reduce revenue by tens of thousands every week.
This is how easy it is for customer revenue to start pouring out of a leaky digital platform. Let’s say you have an event on and you’re promoting it across your app. The background of your product pages changes as a result. There’s an unintended impact that no-one’s spotted: the sizing, colour and add to basket buttons are now lost in the background. Because users can’t see them easily, most don’t persevere. They shrug in irritation and click away to a competitor site – Amazon’s always reliable so they go with what’s easy.
We recently saw an example of this, where one product page bug reduced revenue by £30,000 per week because it affected a high-traffic item. Imagine if you had ten bugs like this concealed in your website. Factor up the costs and you see the scale of the lost revenue. Actively testing for problems like these would flag them earlier and your dev team would be able to fix them quickly, before any customer revenue streams were hurt.
QA best practices for testing at speed and scale
We’ve seen how testing can help your business sustain profits and retain customer spending and we’ve recognised that its bad reputation is underserved. So what’s the best way to approach critical testing positively, to get the maximum benefit and value? Working with many ecommerce brands over the years, we’ve amassed some tips to share.
- Create and maintain the test pack
Take time to create a test pack that focuses on key areas of digital performance and customer experience. As the digital products you’re testing evolve and expand, you’ll need to revisit the test pack to make sure it still has the depth and coverage needed for new features and functions. Build in learning from previous testing cycles to keep the focus on the most critical and risk-prone areas.
- Plan your testing at scale
Don’t skimp on scale if you want to be sure that you’ve covered all bases. All too often, QA leads are asked to release the resources they’ve set aside for testing to upweight urgent dev tasks ahead of the launch deadline. You need to spell out the risks of cutting corners on testing and if your people are being pulled away, make a strong case for extra budget to secure third party testing help.
- Schedule to match your resources and manage expectations
Everyone seems to know the launch date and no-one wants to be the reason for delays to digital features and products that are high profile and promise great results for your business. But you need to plan a schedule that’s realistic. Testing teams are only human – putting them under undue pressure will likely deliver worse results, as they struggle to hit targets rather than focusing on quality outcomes. If the time allowed for testing isn’t enough for the resources you have, you’ll need to deliver unpopular news. Or you could plan to upweight your team with third party resources or use a crowd testing provider that can deliver the equivalent of weeks in days.
- Execute efficiently and quickly with a motivated and experienced crew
In our experience, you get the most accurate results and the best productivity from your test team if you plan a realistic workload and communicate clearly about who’s doing what. Don’t leave it to the last minute: make sure everyone is fully briefed and ready ahead of time. If you’re bringing in third party help, make sure they have the right experience and knowledge for the task. Give them all the information they need and access to the right systems, so they can work as effectively as regular team members.
- Use an agile approach so you can fix as you go
Waiting until all the testing is complete to deliver a set of issues back to the Dev team gives them a headache and creates pressure to fix everything at once. If you adopt an iterative, agile approach to testing throughout the development cycle, and in the final pre-launch stages, you can feed insights back to the dev team as they arise. Testing the robustness of the new fixes will happen continuously, which reduces the risk of unforeseen consequences from lastminute changes.
- Vet your testers’ work
However good your testing team, it’s wise to check their work. Marshalling their findings in a clear framework helps your supervisors identify any anomalies or unlikely results. Checking these out can save your Dev team time and frustration: the more reliable your outputs are, the more confident they will be in the testing feedback they receive, which will help them address issues rapidly and decisively.
- Deliver actionable reporting
Make it easy for the Dev team to understand what the issue is, where to find it and what the preferred outcome would have been. Clear and consistent reporting formats and training your team to articulate problems, supported with evidence (links or screenshots, for instance) can save time, improve understanding and limit the risk of issues being rejected or incorrectly addressed.
A fresh look with the unbiased eyes of the crowd
Crowdtesting can play a crucial role in QA and digital testing programmes. The crowdtester community brings valuable insight into the real user experience because the team comes fresh to your website, application or function. They can help you to grasp how an independent user might view the process and actions they’re expected to use, because they have no idea how the software ‘should’ work from a development perspective. When it comes to spotting obscure issues, crowdtesters really can put your digital assets through their paces, because they explore them in all the ways that a new customer or user might.
Not all crowds are the same though. Choose a crowd provider who can offer you a proven choice of skills, experience, platforms, languages and geographies amongst their testers – whatever you need to replicate the profile of your customers as closely as possible.
A positive partner – put Digivante on your testing team
With years of experience supporting digital QA for high profile brands and ecommerce retail organisations, we’ve got plenty of resources and highly effective approaches that you can rely on for testing at any stage of the SDLC. You can even hand the whole QA process to us, leaving your teams free to innovate new software and improve the existing software behind your apps and websites.
We have experienced test case script writers who have created thousands of test packs for clients. We can keep them up to date with the latest functionality and content too. To keep you on schedule and to help meet tight deadlines, our global community of professional testers can typically embark on test execution with just 24 hours’ lead time.
At Digivante we can test at scale, reducing your test window dramatically. Our testers are a dedicated bunch, collectively working all year round. They’re on call round the clock, because we have testers in every time zone, for a 24/7 service whenever you need it. With agile testing, you receive the results you need fast, which in turn supports rapid fixing and retesting at your end.
Of course, the need for speed is not the only criterion for choosing a testing partner. It’s a given that you need feedback to be accurate. We vet our testers’ work for accuracy and completeness. What’s more, our feedback is actionable. We don’t log defects and pass them to you in a jumble that creates more work in sorting and sifting. Our reports are clear, detailed and specific, backed with evidence.
If you’d like to find out more about improving your testing approach, upweighting your resources with skilled specialist testing leads, accessing a trusted army of vetted crowdtesters, or hiring a fully outsourced QA team, get in touch with Digivante.