The spread of AI tools and technologies within the workplace will have a big impact on enterprise software and how organisations build, buy and rent the systems that support their operations, as we have written about previously. But it could also lead to a new wave of self-built or self-managed platforms.
Foundational models such as LLMs are so capital- and compute-intensive to create that it does not make sense in most cases for organisations to build their own. The same is true for many other components, from natural language processing to automation logic. But the shift from one-size-fits-all enterprise platforms to a more modular approach is creating an opportunity for companies to design their own organisational operating system in ways that were not previously possible.
If you asked leaders a decade ago to describe their organisation’s operating system, they might have spoken about the process landscape, doctrine and culture – and the role of enterprise software was to support the existing operating model, rather than to improve it. But as organisations become more like software, this picture is changing. Processes become services that can be automated or turned into software components. Data and knowledge flows become the lifeblood of operations, rather than management instructions. And the organisation’s service platform becomes the principal system for the coordination of work.
Challenges for vendors
In his predictions for 2025 (yes, it’s that time of year again!), Constellation’s Larry Dignan discussed these dynamics and identified three big shifts in enterprise software that are being driven by AI adoption:
- Marketplaces
- Agentic AI – the new UI for software
- Vendor growth vs. customer budgets
Companies leading in genAI adoption build more than buy. These enterprises are using AI to traverse multiple systems. In a nutshell, genAI and agentic AI will be the user interface for most applications. What’s enterprise software worth if the user interface is redundant?
Salesforce’s Vala Afshar also shared his predictions for 2025, which were unsurprisingly focused on his company’s push for Agentic AI, but there was one point that stuck out for me:
CIOs are now in the business of manufacturing intelligence and autonomous work. They are responsible for creating a work environment where humans and AI agents can collaborate and co-create stakeholder value.”
In other words, CIOs will be leading the development of the future digital workplace where people and AI agents work together.
We might finally be moving away from applying technology to support and solidify old ways of working and towards a situation where we upgrade our organisational operating system and leave behind manual management methods.
Phil Wainewright recently wrote for Diginomica about what enterprises are thinking about building with Agentic AI, and shared this observation:
… doing a better job of introducing existing automations isn’t necessarily the best use of a new technology. As one panelist at Atlassian’s recent event remarked: “We need to challenge the way we’re working today. We shouldn’t just use this AI to give an existing process artificial intelligence, without initially rethinking, is this the way we want to work tomorrow?”
The opportunity to change the way companies create value and get work done is just as exciting for smaller firms and SMEs as it is for those with more substantial technical resources. HBR last week published an excerpt from Karen Mills’ book Fintech, Small Business & the American Dream, which talked about how small firms can leverage existing AI and fintech capabilities to run their business in a smarter, cheaper and better way:
What if this intelligent platform knew when the business might need credit and could offer relevant options which were prequalified? And better yet, what if the platform came with a digital small business assistant who knew all about the business, including the goals and preferences of the owner. What if this bot could respond to requests in plain English to perform daily tasks and improve sales and marketing. Marshaling the predictive power of artificial intelligence and machine learning amassed from data on thousands of business owners in similar industries, what if this bot could help a business owner head off perilous trends and chart a plan for success?
Opportunities for builders
CIO functions have made real progress in recent years towards playing a more integrated role “with the business” and have developed their non-technical skills and competencies. This role is about to get a whole lot more interesting as they coordinate with other stakeholders on topics as diverse as knowledge engineering – to orchestrate the most valuable content that AI agents will need – and workplace psychology in a mixed human-AI workforce.
The digital workplace has evolved considerably since the turn of the century, but it still too focussed on communication rather than orchestrating and coordinating work in many organisations. We published a history of the digital workplace about five years ago that predicted the current moment and what is now possible. As the focus has shifted from intranets to social networks to apps and now to smart platforms and services, the digital workplace has become a stronger part of the fabric of organisations, and now it can become the software that runs them.
Boundaryless just published a guide to designing a platform-based organisation that highlights just how much strategic agility such a model can enable, and highlights the balance that can be achieved between different organisational attributes:
Platform Organizations present complex architectural choices with trade-offs. The key lies in balancing coherence, autonomy, and adaptability, knowing these elements are related by a trilemma and not all optimize simultaneously. … Ultimately, these choices aim to create a structure for rapid innovation, effective resource allocation, and sustainable growth while aligning with the organization’s culture and values.
But even working within existing structures, there is huge scope for improving productivity and reducing management costs by de-bureaucratising traditionally-run organisations, increasing agility and empowering autonomous teams.
I think we are seeing the start of an exciting period of architecting and developing the smart digital workplace, and it looks like there will be more building and integrating where previously firms were mostly buying or renting.
It might sound daunting for some, but the good news is there is so much more that can be done with what companies already have access to. For example, a typical Microsoft 365 instance tends to be over-used for email and meetings, but woefully under-used for knowledge creation, automation and real-time awareness. People don’t even know what it can do, which is a failure of learning, change management and adoption, rather than an inevitable result of the rather fragmented nature of platform. With the addition of co-pilots, LLMs and Azure AI services, it has enough capabilities to become a smart digital workplace, assuming we can break old habits and engage people in working in a more effective way.
After a period of domination by vanilla cloud platforms that made users adapt to their one-size-fits-all workflows, we could see a blooming of smart digital workplace platforms that are a core part of a company’s operating system, and tailored to their own needs and culture. Perhaps then we can treat employee experience with as much care and consideration as customer experience.