CONTRIBUTOR
Co-CEO and Founder,
Launchable

Cut costs, but maintain your productivity. 

That’s been the instruction for engineering and IT leaders over the past few months. Global markets face an uncertain future, and budgets are tightening for companies everywhere. The cloud isn’t getting cheaper, making cloud consumption an easy target for budget cuts.

The do more with less directive can seem like an oxymoron. How can you deliver the expected quantity of releases while shrinking your cloud consumption? Two big components of cloud spend are running your production environments and running your testing. Given that taking production offline is not in question, how can you maintain your quality standards if you reduce your testing? How can you maintain your quality standards with less room for testing?

Maybe, just maybe, there’s a way for engineering and IT leaders to strike a balance. 

Your company, like many others, probably isn’t immune to the pressures of economic uncertainty. 81% of IT teams have been directed by their C-suite to halt or reduce their costs. If budget cuts haven’t hit yet, they’re likely on their way; some experts say we’re headed down for a little longer.

C-Suite executives have updated their views on cloud services amidst economic uncertainty. A couple of years ago, 42% of CIOs and CTOs said security investments were their top cloud priority.

Fast forward, and priorities look more like this: Aligning the speed of IT delivery with speed of business (25% of executives), ensuring compliance needs are met (22%), and increasing direct revenues (20%). An emphasis on speed, cost savings and compliance has changed the equation for DevOps and IT. 

We’ve replaced the “releases at all costs” mentality with a more measured approach — while still being evaluated on the quality and quantity of releases. The most successful engineering and IT leadership will take simple yet effective steps: Prioritizing cloud migrations, looking inward to find inefficiencies, failing faster and smarter and investing in data-driven testing solutions.

In uncertain times, striking the right balance requires creativity and strategic tweaking.

The Cloud Cost & Consumption Dilemma for Engineering & IT Leaders

Let’s say you have a test pipeline that’s running on the cloud. You’re getting pressure from your boss to reduce your cloud consumption to align with budget decreases.

You’ll have a few options:

  • Test on bigger machines that move much faster & reduced execution times (but have higher costs per machine hour)
  • Parallelize your tests on multiple machines & reduce execution times (which will increase cost as you pay for your own machine)
  • Bring your dev and production environment back on-premises

Each of your options is flawed. That’s not to mention that cloud costs for services like AWS are spiking, with some companies doubling their storage rates.

Some teams are discussing bringing development and production back on-premise. It is not an easy option as most teams have spent significant resources to move to the cloud. This option isn’t turn-key, as the migration back has to be planned in addition to allocating budget for capital expenditure.

4 Ways To Reduce Cloud Spending & Consumption in 2023

Here’s what the smartest organizations are doing to achieve these results despite the economic rainclouds that are passing by.

1.  Prioritize to get the biggest bang for your cloud buck.

Cloud comes at a premium these days. Let’s come back to the C-Suite’s main goals for cloud investments in 2023 and beyond: speed, savings and compliance. Some orgs might assume slowing cloud migrations will slow the budget burn, but it’s the opposite.

“A common misconception is that organizations can reduce costs by slowing down cloud migrations and working within their on-premises environments, which they’ve already paid for. However, compared to cloud environments, on-premises data centers require continuous operating support in the form of labor, utilities, leases, and licenses to maintain systems, manage refresh cycles, and combat outages. Cutting costs in any of these areas can also lead to expensive issues.” – McKinsey Digital

Increase your spending efficiency by prioritizing the workloads that generate the most value for the business. Behind them in the queue will be workloads with the highest overhead or ones that are running on the most outdated equipment. That way, you’re prioritizing valuable cloud space for the systems that need it the most.

2.  Find the lowest hanging fruit.

Not every cost-savings measure has to be drastic. Sometimes, a collection of small measures can help you fit into your budget.

McKinsey told the story of one major public-sector agency, which achieved roughly 20% cloud savings by adjusting cloud services to better match application needs, disposing of assets it was paying for but no longer using, instituting basic guidelines for tiered storage, and updating the instances to the latest version.

“Releases at all costs” can be reckless; small changes amount to a more measured cloud approach.

3.  Fail faster – but fail smart.

This tenet can be applied across your department. Experiment with simple fixes to your cloud services, and apply the most effective ones to other environments.

On the testing front, it’s smarter to fail fast and triage flaky tests as soon as possible – it’s the lowest hanging fruit to tackle and lightens more strenuous testing later in your pipeline. UI tests often are the most expensive, but the right data-driven software can help you reduce test execution times. One of the largest cloud computing companies in the US took this approach and reduced their test execution time by 90%, cutting cloud costs for test executions. 

The key for engineering and IT managers is shortening test suite runtimes. If this can be achieved, you’ll save cloud consumption and developer hours, giving you more output from the same developer team. You’ll also find failures faster, meaning you’ll have fewer red builds, faster releases and happier, more fulfilled developers.

4.  Invest in data-driven software testing.

Data-driven testing software uses machine learning to generate the most insights out of the data you have within your existing dev pipeline.

Without using data to assess your pipeline, you’re likely spending both time and cloud infrastructure costs testing issues that never affect your ability to launch, and wind up focusing on what really matters at the very end of your pipeline. That’s not conducive to saving cloud consumption and spending.

The right software ranks tests by importance to code changes and allows you to create unique subsets in real time. That means you can run a fraction of your test suite while still maintaining high confidence that if a failure exists, it will be found.

Final Thoughts & Takeaways

No one said shrinking budgets were fun, but they don’t have to slow down your development pipeline. Adopting a new, more-measured approach — one that’s augmented by smart AI tools can help engineering and IT leaders maintain quality and quantity amidst new challenges.