Overcoming the conflict between speed and quality

Today’s winners are defined by speed. If you can develop an idea, turn it into a tangible product, and get it out to your target audience faster than the competition, you are in a much better position to win. It doesn’t matter how great your offer is; it’s much harder to win customers if you’re always trailing behind competitors. 

Pressure to deliver

It’s an ethos that underpins an entire philosophy in the start-up world: the minimum viable product. Read any best-practice guide, and there will likely be a section on getting an MVP out to customers. The same guide will almost certainly have stories of companies and individuals with incredible ideas no one has ever heard of. Why? Because they weren’t quick enough. 

This isn’t something confined to start-ups, though. The focus on speed is one organisations of all sizes have, with getting applications into the hands of users a priority. 

It might be something internal for employees or something external for customers. Whatever the intended audience, the aim will be to help them engage digitally with the business by creating new, enjoyable experiences via the app. 

That puts pressure on organisations to keep churning out apps, updates, and new releases. Yet while early prototypes and new products can be MVPs, there comes the point where user expectations are such that they expect both speed and quality. Anyone that’s ever looked at the reviews of a website or web based application on a store or marketplace will have seen comments complaining about new versions not working correctly, or updates that haven’t fixed existing issues. 

These expectations aren’t limited to customer facing applications either (e.g. ecommerce stores or front-end website); we’re at a point now where we all expect digital services to just work, including internal business applications or account restricted service portals that rely on data, workflows and complex logic. 

The conflict between speed and quality

But when we also all expect things quickly, that creates tension. Delivering services at speed that meets quality standards is possible; it is also a massive challenge for many organisations. 

Part of the reason is where companies are in their automation journey. Very few, if any, are fully automated. Even in technology teams, some processes are manual. What that causes is a system running at multiple speeds. Suppose you’re deploying Continuous Integration and Continuous Development (CI/CD) or DevOps. In that case, you might well have hugely accelerated your development capabilities and, in doing so, be creating and deploying services to meet the needs of your users rapidly. But if your testing approach is still manual, you’ve got operations moving at different rates. Either you have a build-up of new products waiting to be tested, in which case users go waiting, or you have to decide how you test (do you test everything but not in great detail or only specific services) and accept quality as a whole will take a hit. 

To be fair, testing has long been a challenge in software development. It’s absolutely required but by its very nature is the antithesis of the agile, continuous approach most organisations try to follow when developing new products and solutions. 

Time to truly automate testing

Today, it’s only got worse. Testing teams are under pressure to test faster so that offerings can get into the market quicker. And as companies increase their use of automation, particularly in areas like DevOps, the problem will only worsen unless companies start automating their test. 

For some, that’s a none-starter. Their view is that the value of testing comes from having people, with all their experiences, faults, and insights, check that digital services work for the user (as well as meet other requirements). 

In the past, they were right. Technology simply couldn’t replicate what people could do. Businesses that deployed coded automation testing spent as much time fixing problems in the testing as they did with manual testing, rendering the whole exercise pointless. In such a situation, it’s not surprising that many still shy away from the concept of automating testing. 

The problem is budgets aren’t available to hire enough resources to meet the demands of businesses that want digital services deployed at accelerated rates. Nor is the desire to manage large teams – the overhead required, including having to continually train and deploy new hires when people come and go… it’s too much to scale.

Improving productivity and delivering value back to the business

However, technology offers a solution. We’re entering the age of hyperautomation: the combination of robotic process automation, natural language programming, artificial intelligence, machine learning, and codeless tools. One of Gartner’s top strategic trends for 2022, hyperautomation allows businesses to identify and automate processes on a scale previously unobtainable. 

It’s being deployed in various settings and operations, including development. That means it also needs to be applied to testing as a matter of course. There’s no point in accelerating the creation of digital services to get them to market quicker if they then have to go through manual testing processes. 

With hyperautomated testing, companies can rapidly increase the amount of code they can run through quality assurance processes, removing bottlenecks. Whether it’s the challenges posed by dynamic applications, testing silos, maintenance, lack of resources, or low coverage, this new approach to automated testing allows businesses to scale their testing teams. Rather than working on one application or project at a time, testers can now simply monitor multiple projects while NLP, AI, and ML assess, learn and apply findings to the testing process. This isn’t about humans being removed– there still needs to be oversight monitoring the tests themselves – but it allows small teams to accomplish much more. And that is the key. Use technology to empower and scale vs replace.

What’s more, taking the day-to-day running of existing testing away from QA teams gives those employees more time to focus on more complex problems, as well as scaling testing with additional, valid tests that cover more the application. Both testing productivity and delivering additional value to the organisation increase.

Overcoming the tension between speed and quality

Companies that fail to move fast are destined to struggle. At the same time, users expect differentiated experiences, which can only be achieved by deploying high-quality digital services. This tension is exacerbated in today’s marketplaces, but just as more processes are being automated to meet the demands for speed, so can that technology be applied to testing. Forward-thinking organisations that deploy hyperautomated testing approaches will be the ones that make things fast without compromising on quality. 

James Bent

James Bent, VP of Solutions Engineering, Virtuoso

Unlocking productivity and efficiency gains with data management

Russ Kennedy • 04th July 2023

Enterprise data has been closely linked with hardware for numerous years, but an exciting transformation is underway as the era of the hardware businesses is gone. With advanced data services available through the cloud, organisations can forego investing in hardware and abandon infrastructure management in favour of data management.