Innovation isn’t just about ideas, it’s about momentum. But it’s hard to maintain the drive when working days are filled with administrative tasks, meetings and other interruptions.
According to a recent report from Miro, for every hour spent on strategic or creative work, knowledge workers spend three hours on administrative tasks, meetings and communications. Six in 10 workers say these momentum blockers sap their concentration, reduce their creativity and raise their stress levels. Teamwork and collaboration suffer too: 70% of respondents experience collaboration issues at least monthly, and 55% said projects and initiatives at their company often lose momentum.
Excessive busywork is also tied to organisation-level problems, such as data and communication siloes. Constant emails and messages pinging across different channels and teams can create communication chaos. Ceaseless documentation and reporting tasks, and information that sprawls across different tools and apps, also leads to duplicated work, missing data and context and slower decision-making.
AI promises to address many of these issues and create more time for the momentum work that drives innovation. It presents an opportunity for organisations to fundamentally rethink how work gets done, which could improve both the wellbeing of employees and teams’ ability to innovate at speed. But it isn’t a silver bullet. Poor implementation can compound many issues with existing working practices.
“It’s not about taking a new technology and tacking it on to existing processes,” says Grisha Pavlotsky, chief transformation officer at Miro. “It’s an opportunity for leaders to ask: ‘Is my operating model right for this new era? Are processes and roles still valid?’”
Improving teamwork
Major technology shifts of the past, such as electricity transforming how factories operated, highlight the need to rethink work rather than incrementally improve it. Yet many recent implementations of AI in the workplace have focused on individual productivity gains in the strictures of current processes and roles.
Employees have gained the ability to code faster with co-pilots or produce a presentation in seconds, for example. But this solo-focused approach to AI neglects the more profound benefits it offers for teamwork and collaboration.
“Instead of asking, ‘How do I give copilots to every employee?’ companies should be asking whether their organisational design, their jobs-to-be-done, their processes – having this person doing this job and then handing it over to this person – are right,” Pavlotsky argues.
Organisations that fail to ask these questions risk thinking they’re moving faster while not actually progressing much at all. They may achieve high AI-adoption rates, for example, yet see only relatively small productivity improvements at the team and organisational level.
For every hour spent on strategic or creative work, knowledge workers spend three hours on administrative tasks
“You need to give AI to the team, not to individuals,” says Pavlotsky. “If you miss that nuance, you will be fooled by the fact that AI adoption is very high in your organisation. Yet in reality you’re not going to be identifying the right problems to solve, and you’re not going to be experimenting fast enough.”
Organisations often see AI as a means to speed up the delivery phase of product development. But Pavlotsky argues that many of the most transformative opportunities lie earlier in the development life cycle, during the discovery and definition phase.
Indeed, in a world where the cost of building software is plummeting thanks to AI, the ability to iterate rapidly on concepts, test assumptions quickly and reach high-quality specifications faster is becoming a key competitive battleground.
“We’ve figured out how to optimise the delivery side of product development,” says Pavlotsky. “Tools like GitHub Copilot and Codex have solved that. So the key question now is not, ‘Can I build something?’ It’s, ‘Should I?’”
AI tools can synthesise data and identify the key insights, brainstorm ideas for improvements and even prototype a new user experience to generate feedback. These AI sidekicks – intelligent agents that work alongside teams – can even provide instant access to specialised expertise by emulating the response of a senior leader, helping teams move through the early stages of product development at greater speed. Ultimately, they should be able to iterate and test many more ideas, enabling them to reach product-market fit much faster.
However, none of this is possible without AI that has team-level shared context. “If everyone uses their own AI individually, then in meetings you will just argue about prompts and contexts. ‘What did you feed the model? What version of the file did you use?’ That’s wasted time and the loudest voice may dominate, leading to misguided decisions.”
The right context
Context engineering is an important new discipline that addresses this issue. It’s about ensuring AI has the right shared, dynamic inputs. Because just as humans need shared context when they solve the problem, team-level AI also needs it to provide consistent, high-quality outcomes.
Team-level AI that draws input from a collaborative workspace can also serve every member of the team with outputs in their preferred form factors: technical diagrams, documentation, data tables, slides, prototypes and so on. But while the outputs may differ, all of them are based on that same shared context.
Crucially, everyone must be able to see and curate this evolving context in real-time. “Most current tools let you upload files that are static at that moment in time,” says Pavlotsky. “But if you’re working on Miro canvas, you have a lot of context there that the AI is going to be using for a task. It’s live, it’s dynamic and it’s visible, so everybody on the team understands what the AI is using. Furthermore, Miro AI can ingest context in all types of form factors: visual, docs, tables, dashboards and diagrams.”
Smarter use of AI at the team level will also enable human workers to shift from specialists focused on narrow domains to orchestrators of end-to-end value creation. However, this requires an understanding of how different work streams connect, what intermediate outputs feed into final deliverables and how quality is maintained across complex workflows.
“You’re not just responsible for one intermediate artifact – you’re responsible for the final artifact and you need to understand what it takes to get there,” says Pavlotsky.
While this shift will require new competencies – and leaders to have a clearer understanding of the value of certain artifacts – it could also help to free workers from the confines of their current roles, which are far too focused on maintenance work today.
The organisations that thrive in the AI era won’t be those that simply deploy AI tools to help individuals carry out existing – and often draining and unfulfilling – tasks more efficiently. They’ll be the ones that use game-changing technology to fundamentally rethink how work gets done.
Innovation isn't just about ideas, it's about momentum. But it’s hard to maintain the drive when working days are filled with administrative tasks, meetings and other interruptions.
According to a recent report from Miro, for every hour spent on strategic or creative work, knowledge workers spend three hours on administrative tasks, meetings and communications. Six in 10 workers say these momentum blockers sap their concentration, reduce their creativity and raise their stress levels. Teamwork and collaboration suffer too: 70% of respondents experience collaboration issues at least monthly, and 55% said projects and initiatives at their company often lose momentum.
Excessive busywork is also tied to organisation-level problems, such as data and communication siloes. Constant emails and messages pinging across different channels and teams can create communication chaos. Ceaseless documentation and reporting tasks, and information that sprawls across different tools and apps, also leads to duplicated work, missing data and context and slower decision-making.
