As 2019 drew to a close, DevOps practitioners and vendors were reflecting on a year dominated by hyphenated variations of the term — DevSecOps, GitOps, NoOps — and asking what would come next. A range of industry experts shared their predictions for 2020 with DevOps Online. Many of their forecasts proved accurate, and several remain directly relevant to engineering teams today. Below is a structured summary of those predictions, organised by theme.
IoT and Embedded Video Will Expand the Attack Surface
Malcolm Murphy, Technical Director EMEA at Infoblox, highlighted video-capable IoT devices as an emerging security concern.
“I strongly suspect that video cameras in IoT connected devices, such as video doorbells and embedded cameras in smart speakers, will be the next big threat to the network due to the richer information sets they hold. Since these devices are collecting more and more data, they are becoming increasingly interesting targets for hackers, so that data is at a higher risk of being compromised. Despite this risk, security is still not being prioritised by manufacturers in the design of smart products, so until this is addressed, we’ll continue to see breach levels rise.”
Malcolm Murphy, Technical Director EMEA, Infoblox
The prediction reflects a persistent gap in IoT security: hardware manufacturers have historically deprioritised security in favour of speed-to-market, leaving network defenders to compensate downstream. This remains a live concern for DevSecOps teams integrating smart devices into enterprise environments.
IT Will Need to Govern Citizen Developers
Jennifer Gill, Senior Director of Product Marketing at Pegasystems, foresaw the governance challenges that low-code adoption would create.
“Low-code platforms have been touted for their ease of use so business professionals can develop their own applications. But without IT’s involvement, applications can go rogue and cause governance, maintenance, and usability issues. To ensure citizen developers don’t go too far astray, IT departments will utilise low-code platforms with templates to keep business users within the guardrails of corporate governance.”
Jennifer Gill, Senior Director Product Marketing, Pegasystems
The tension between empowering non-technical users and maintaining platform integrity has only intensified since 2020. Platform engineering teams now frequently publish internal developer portals and golden-path templates for exactly this reason — providing guardrails without blocking productivity.
Automotive OEMs Will Embrace Mobility-as-a-Service
John Phillips, MD EMEA at Zuora, predicted that Mobility-as-a-Service (MaaS) would push automotive manufacturers toward subscription-centric, customer-focused models.
“In 2020, we will see automotive OEMs use this knowledge and pair it with technology that will help them integrate their services with other suppliers both inside and outside of the auto supply chain in order to realise the full potential of Mobility-as-a-Service. The true winners in MaaS will place customer relationships at the centre of their businesses.”
John Phillips, MD EMEA, Zuora
From a DevOps perspective, this prediction foreshadowed greater demand for continuous delivery pipelines capable of supporting over-the-air (OTA) software updates, API-driven integrations with third-party mobility providers, and real-time telemetry pipelines for subscriber data.
Open Standards Will Underpin Agile DevOps at Scale
The Open Group argued that as organisations scaled their DevOps practices, open standards would become critical for maintaining consistency across complex value chains.
“Open standards, including information models such as the IT4IT Reference Architecture, can provide a framework for agile development, helping to enable the standardisation of components and interfaces, and allowing teams to seamlessly interchange information on product delivery and use without disrupting the entire value chain.”
The Open Group
This aligns with the broader industry movement toward standardised software supply chain specifications, such as SLSA (Supply-chain Levels for Software Artifacts) and SBOM (Software Bill of Materials), both of which gained significant traction after 2020.
DevOps Will Move Up the Value Stream
Mark Levy, Director of Strategy for Software Delivery at Micro Focus, predicted that DevOps would expand beyond engineering teams to encompass portfolio and product management.
“In 2020, there will be a greater focus on connecting portfolio/product management with the other practices of DevOps. Fast feedback should not only be in the boundaries of Dev and Ops but encompass the entire value chain. DevOps starts to move up the value stream to provide better alignment with the business.”
Mark Levy, Director of Strategy – Software Delivery, Micro Focus
This prediction maps closely to what became known as value stream management (VSM) — a discipline that gained significant adoption as organisations sought to measure and optimise the flow of work from business idea to production deployment.
DevOps Will Mature Into a “So What” Topic
Edge Testing offered a contrarian view: that the hype surrounding DevOps would dissipate as the methodology matured into standard practice — much as Agile had done before it.
“DevOps will go the same way as Agile in that it will become a ‘so what’ topic. Agile is now a fully matured methodology with the majority of testers having some exposure which they can put on their CVs. DevOps will follow this route. Although we will continue to see one new tool or another being launched weekly to bolt onto the DevOps tool suite, the main set of tools are maturing nicely and the skill sets to support these in the industry are growing.”
Edge Testing
This proved largely accurate. By the mid-2020s, DevOps had become a baseline expectation for engineering roles rather than a differentiator, and attention shifted toward more specific disciplines such as platform engineering, SRE, and DevSecOps.
The CISO Will Become a Full Member of the DevOps Team
Malcolm Isaacs, Head of ADM Solutions Marketing at Micro Focus, predicted the full integration of security leadership into DevOps pipelines — what the industry now calls DevSecOps or shift-left security.
“Just as DevOps strives to deliver value quickly to the customer, it has the potential to unintentionally introduce security vulnerabilities quickly as well. Over the course of 2020, any remaining barriers between CISO staff and DevOps teams will be broken down, with CISO staff becoming full-fledged members of the DevOps team. Security will no longer be a bolt-on activity, and will become a standard component of any DevOps pipeline.”
Malcolm Isaacs, Head of ADM Solutions Marketing, Micro Focus
High-profile supply chain incidents — including the SolarWinds compromise in late 2020 — dramatically accelerated this shift and made security integration in CI/CD pipelines a regulatory and business imperative rather than a best-practice aspiration.
Open Source Business Models Will Need to Reinvent Themselves
Perry Krug, Architect at the Office of the CTO at Couchbase, identified the growing tension between open-source projects and cloud providers that commoditise their distributions.
“In 2020, vendors will have to realise that the Open Source game has changed. They must instead see themselves as service providers, from management to support and other value-adds, in order to ensure that they can offer something that others cannot. The business model is to innovate — where cloud providers won’t — and to provide better and more differentiated service, not just differentiated software.”
Perry Krug, Architect – Office of the CTO, Couchbase
This tension played out publicly with licensing changes from projects including Elasticsearch, HashiCorp Terraform, and MongoDB, all of which adopted more restrictive licences to limit cloud-provider free-riding. The debate around open-source sustainability continues to evolve.
“No-Ops” Will Die — Ops Will Evolve
New Relic argued that the concept of eliminating the Ops function was misguided, and that automation would elevate Ops professionals rather than replace them.
“Next year, more organisations will start to ditch the notion of ‘No-Ops’ and recognise value in their Ops team members. Companies will always need Ops people, but as technology and automation processes mature, Ops professionals will be able to work on higher-quality problems.”
New Relic
The rise of Site Reliability Engineering (SRE) as a formal discipline bears this out. Rather than eliminating Ops, organisations increasingly embed SREs within product teams to own reliability, scalability, and observability — exactly the higher-order problems New Relic described.
Disruptors Will Set the Pace for Customer Engagement
David Parry-Jones, VP EMEA at Twilio, predicted that API-driven communications platforms would push traditional businesses to match the customer engagement capabilities of digital-native disruptors.
“With the advent of APIs and cloud-based communications platforms, customers can now be easily reached across a multitude of different channels according to their preference, and call-time vastly reduced through the application of AI. In 2020 more organisations should begin to ask — why can’t we do this too?”
David Parry-Jones, VP EMEA, Twilio
For DevOps teams, this translated into demand for event-driven architectures, API gateway implementations, and real-time notification pipelines — all areas that saw rapid tooling investment in the years following this prediction.
Data-as-a-Service Will Become a Revenue Stream
Alberto Pan, CTO at Denodo, predicted that organisations sitting on proprietary datasets would begin packaging and monetising that data for third parties.
“Since many big companies will now own high-value, unique data and services, the next logical step is reusing this infrastructure in order to offer them to third-parties. This trend will be significantly accentuated during 2020 in all major industries. From the investment standpoint, this will involve higher demand for the technologies involved in creating and exposing data as a service, like GraphQL, Data Virtualisation and API management tools.”
Alberto Pan, CTO, Denodo
The prediction preceded a significant expansion in data mesh architectures, data marketplaces, and regulatory scrutiny of commercial data sales — particularly around location and behavioural data. DevOps teams in data-intensive industries increasingly own pipelines that serve external as well as internal consumers.
IT Will Reclaim Its Strategic Influence
Brian Kelly, CEO at CloudBolt, closed the predictions roundup with a broadly optimistic view: that IT practitioners — not just CxO-level decision-makers — would drive tool selection for hybrid cloud management.
“The IT community (not the obvious power players) will determine the winners of the go-to tools for managing hybrid cloud infrastructures, as they seek extensible, easy-to-deploy, cloud-native solutions.”
Brian Kelly, CEO, CloudBolt
This bottom-up adoption pattern is well documented in the DevOps toolchain space — platforms like GitHub Actions, Terraform, and Kubernetes gained enterprise foothold primarily through practitioner adoption before receiving top-down endorsement.
Key Themes Worth Tracking
Looking across these predictions, several recurring themes emerge that remain relevant for DevOps and platform engineering practitioners:
- Security shifts left — and becomes shared: DevSecOps is no longer optional. CISO involvement in pipeline design is increasingly standard practice.
- Ops evolves rather than disappears: Automation raises the floor for operational work; it does not eliminate the need for operational expertise.
- Open-source sustainability is unresolved: The conflict between open-source ideals and commercial cloud distribution is ongoing and actively reshaping licensing strategies.
- Citizen development requires governance: Low-code proliferation without platform guardrails creates technical debt and compliance risk.
- Value stream visibility matters: Fast delivery is only valuable when aligned with business outcomes — a principle central to modern platform engineering and DevOps measurement frameworks.
Original predictions sourced from DevOps Online, published December 2019. Contributor affiliations reflect their roles at time of publication.