Fluid Teams, Fragile Belonging: Skills, Teams and the Shifting Unit of Work

Skills marketplaces are having more than a moment. Internal talent marketplace adoption rose from 25% of organisations in 2024 to 35% in 2025, according to SHRM (Society for Human Resource Management). The logic is compelling: reorganise work around verified capability rather than job titles, let talent move fluidly to where it's needed, and make career progression a function of skills visibility rather than tenure. There is a growing pool of research that points to evidence supporting the logic. Deloitte's research on internal talent marketplaces (‘Activating the Internal Talent Marketplace’) found that Unilever redeployed more than 8,000 employees and 300,000 hours of work through its FLEX Experiences marketplace during the pandemic, demonstrating the model's capacity for rapid, large-scale redeployment.

McKinsey's research presents compelling data using different metrics: in a survey of more than 1,800 executives, the talent practice most predictive of outperforming competitors was frequent reallocation of high performers to the most critical roles. Companies that did this quickly were 2.2 times more likely than slower reallocators to outperform on total returns to shareholders. Unilever's experience supports this reporting a cumulative 41% improvement in productivity and more than 700,000 hours of capacity unlocked since it’s FLEX Experiences launch, alongside a 20% increase in internal collaboration time.

Why do companies decide to adopt these platforms in the first place? Deloitte's interviews with 13 organisations running live marketplaces found three distinct stated purposes. Just over half framed the marketplace primarily as a deployment tool — matching skills to work faster to improve productivity. About half framed it as a mobility and career-development tool, aimed at giving employees visibility into internal moves and improving retention. A quarter described something more ambitious: an evolving ‘future of work’ model meant to combine talent, career, engagement, and innovation across the whole organisation. Many companies start in the first category and drift toward the third as the marketplace matures.

Beneath these stated purposes, the same reasons recur across the wider literature: avoiding the cost of external hiring, responding quickly to shifting demand, making idle skills visible to managers who don't know they exist, retaining people who would otherwise leave for a role elsewhere, and reducing the network-and-pedigree bias that informal staffing tends to reproduce. Worth noting is the research published in Administrative Science Quarterly that found that employees who moved internally into a role significantly outperform external hires in their first two years.

Given all of the stated intent, Gartner's most recent research (October 2025) found that despite continued investment in marketplace technology, internal mobility rates have begun to flatten. Gartner's own framing is candid: one in five employees will need to be redeployed by 2030 as roles change, and organisations aren't prepared for it. So, what could be stalling the rate of uptake in marketplaces?

Most of the debate about skills marketplaces focuses on the technology and the taxonomy. Can you map skills accurately enough? Can the data be trusted? Can you get past managers' resistance to use them?

But could other forces be at play that dilute the appeal of making teams fluid?

Consider this scenario. A firm dismantles its fixed project teams and moves to a skills marketplace. High performers get pulled into high-priority projects faster. Utilisation improves and talent data is finally measurable and accurate. Eighteen months in, three of the highest-potential performers leave. Exit interviews say the same thing: 'I’m burnt out and didn't feel like I belonged anywhere'.

What went wrong? And was it avoidable - or the price of the model?

Decades of team dynamics research - Hackman, Edmondson, Google's Project Aristotle - tells a consistent story about what drives high performance in teams. According to this research, it is not primarily a function of individual skill quality. It is a function of relational conditions: psychological safety, role clarity, shared mental models, and trust built through repeated interaction over time.

Project Aristotle is the most rigorous data point. Google studied 180 teams over two years, examining 250 different team attributes. What they expected to find was that team quality was a composition problem - get the right people, the right mix, the right expertise. What they uncovered was that the dominant variable was not who was on the team but how the team worked together. Psychological safety - the belief that you can speak up without risk of humiliation or punishment - was the single most important factor. The variables that didn't predict performance at all included co-location, individual IQ, and team size.

Here is where the tension with skills marketplaces emerges. This research was conducted on stable teams. Teams that knew who was in them and where trust had time to compound through repeated interaction. The skills marketplace model, at its extreme, creates the opposite conditions: teams that assemble around a project and disperse when it concludes. The skills match may be excellent. The psychological safety may never develop.

It's important to be precise about which version of the model this applies to. Redeploying someone into a stable, ongoing role doesn't disrupt team continuity - it's closer to smarter internal hiring. The tension explored here is specific to the fluid, project-based version of the model: teams that assemble around a piece of work and disperse once it's done, with people cycling through several in sequence. Unilever's FLEX Experiences, the example introduced above, is built in this way.

Organisational researchers Wendy Smith and Marianne Lewis offer a useful lens here. Their work on paradox theory identifies recurring tensions that organisations don't resolve so much as learn to hold simultaneously - competing demands that are logical on their own terms but pull in opposite directions when combined. Two of these are directly at play in the skills marketplace model. What they call the organising paradox is the tension between the flexible, fluid structures a marketplace needs to function and the stability and coordination that high-performing teams require.

What they call the belonging paradox is the tension between an individual's need for a stable group identity and the challenges of moving continuously between teams and projects. A skills marketplace, by design, intensifies both at once: it asks the organisation to be more fluid in how it structures work, and it asks individuals to find belonging inside that fluidity. Smith and Lewis's central argument about paradoxes is that tensions aren't problems to solve once and move past - they're ongoing, and the organisations that manage them well are the ones that build the capacity to hold both sides at once.

This isn't an argument against skills marketplaces as a way of organising work. It's an argument for getting clear about what the model requires to work.

It also points to who has to do that work. Every organisation Deloitte interviewed for its marketplace research named manager engagement as the single biggest factor in whether a marketplace succeeds - ahead of technology or process. A related Deloitte finding, drawn from its 2019 Global Human Capital Trends research, put a number on the resistance side of that: roughly 46% of managers actively resist internal mobility, largely out of anxiety about losing people they can't quickly replace. That resistance is a symptom of asking leaders to do something most of them have never been trained to do: build trust and belonging in a team that reconstitutes every few months, made up of people who've never worked together before.

Harvard's Amy Edmondson, whose research anchors the psychological safety findings above, has a name for this specific capability: teaming - the active, ongoing work of coordinating and building trust on the fly, without a stable team structure to fall back on. Her argument is that teaming doesn't happen naturally; it has to be actively led, and relearned, for every new configuration. Leading a team that reconstitutes every few months is recognised as a distinct capability, and one that traditional leadership development, built around stable teams, hasn't caught up to.

Fluid team structures demand more deliberate investment in the conditions for psychological safety - and that investment doesn't end once the structure is in place. Belonging isn't an emergent property of putting capable people in a room together; it's something that has to be built and rebuilt, repeatedly, often in compressed timeframes and across people who haven't worked together before. To do this well means not treating team belonging as an afterthought but as an ongoing practice, sustained at the same pace as the fluidity it's meant to counterbalance: onboarding rituals for new project teams, embedding explicit norms for how decisions get made, structured mechanisms for surfacing disagreement early.

Belonging, once established, doesn’t necessarily hold. The exit interview data from the scenario above isn’t hard to imagine. People can tolerate a lot of structural ambiguity if they feel they belong somewhere. Remove that, and capability matching alone is rarely enough to keep a high performer, and the capable generally have the most external options.

The skills marketplace model doesn't cause belonging failure on its own. Neglecting the tension does - treating it as a box to tick once, rather than a balance to keep holding. There's no version of this model where the paradox goes away: the more fluid the organisation becomes, the more deliberate it must be about creating belonging, as a discipline that runs for as long as the marketplace does.

Sources

SHRM (Society for Human Resource Management). 2025 Talent Trends / Recruiting Trends research on internal talent marketplace adoption.

Gantcheva, Ina, et al. 'Activating the Internal Talent Marketplace.' Deloitte Insights, 17 September 2020.

Barriere, Mike, Miriam Owens, and Sarah Pobereskin. 'Linking Talent to Value.' McKinsey Quarterly, 12 April 2018.

Unilever and Gloat. FLEX Experiences case study reporting (productivity, hours unlocked, and internal collaboration figures).

Bidwell, Matthew. 'Paying More to Get Less: The Effects of External Hiring versus Internal Mobility.' Administrative Science Quarterly, vol. 56, no. 3, 2011, pp. 369–407.

Gartner, Inc. 'Gartner Identifies Four Trends Talent Management Leaders Should Prepare for in 2026.' Press release, 29 October 2025.

Duhigg, Charles, and Google re:Work. Project Aristotle research on team effectiveness (180 teams, 250 attributes studied over two years).

Hackman, J. Richard. Leading Teams: Setting the Stage for Great Performances. Harvard Business School Press, 2002.

Edmondson, Amy C. 'Psychological Safety and Learning Behaviour in Work Teams.' Administrative Science Quarterly, vol. 44, no. 2, 1999, pp. 350–383.

Smith, Wendy K., and Marianne W. Lewis. Both/And Thinking: Embracing Creative Tensions to Solve Your Toughest Problems. Harvard Business Review Press, 2022; and 'Toward a Theory of Paradox: A Dynamic Equilibrium Model of Organizing.' Academy of Management Review, vol. 36, no. 2, 2011, pp. 381–403.

Volini, Erica, et al. 'Talent Mobility: Winning the War on the Home Front.' 2019 Deloitte Global Human Capital Trends, Deloitte Insights, 2019.

Edmondson, Amy C. Teaming: How Organizations Learn, Innovate, and Compete in the Knowledge Economy. Jossey-Bass, 2012.

Driskell, Tripp, Gregory Funke, Michael T. Tolston, August Capiola, and James Driskell. 'Supporting Fluid Teams: A Research Agenda.' Frontiers in Psychology, 2024, DOI: 10.3389/fpsyg.2024.1327885.

Thomas, Christopher H., Foster Roberts, Milorad M. Novicevic, Anthony P. Ammeter, and Dragan Loncar. 'Familiarity and Fluid Team Performance: Leadership and HRM Implications.' Research in Personnel and Human Resources Management, vol. 36, Emerald Publishing, 2018, pp. 163–196.