Many technology companies have converged on an operating model where Agile teams own most product and engineering work, while management focuses on resolving issues at the seams. When alignment breaks down, decisions escalate up the hierarchy. This is a rational response to a practical reality: Small teams need the autonomy to move quickly and solve problems, even as a complex portfolio of products must ultimately work together.

This model isn’t wrong, but it shouldn’t be where the conversation stops. Tech operating models have always evolved—and they continue to do so. Bain’s latest research shows this evolution is ongoing, and some leaders are moving beyond this baseline, adopting new approaches that unlock faster growth and sustained innovation.

Sustained growth remains elusive

To understand more clearly how companies work, we used a Bain proprietary AI tool to analyze about 300 companies. The solution combines traditional indicators with public knowledge of the organization to build a view of how operating systems behave in practice.

While fast growth and innovation are normal in the tech industry, sustained high growth over many years and repeated innovation beyond the original core is far less common (see Figure 1). Our research found that only 33 of 290 tech companies that we analyzed showed consistent growth of at least 20% per year for 10 years in either revenue, market capitalization, or total shareholder returns. Only 9 of these 33 added at least two successful adjacencies beyond their core.

Figure 1

These standout performers organize and operate differently. First, their organizational structures are flatter (see Figure 2). Second, their leaders and employees are much more likely to exhibit what we call a Founder’s Mentality®—where the front line thinks like owners, leadership is obsessed with the front line, and everyone acts like an insurgent.

Figure 2

More significantly, these top performers show greater diversity in their choice of operating models (see Figure 3). While the seam-focused model (boundary enforcement) is still common, three other common models help them expand into new businesses:

• Habit cultivators reinforce decision habits and norms. For example, one online service provider has a defining habit of turning disagreements into A/B tests with clear decision criteria, the outcomes of which become the decision record. Teams are rewarded for reducing uncertainty, not winning debates—resulting in fewer escalations at the seams.
• Authority weavers delegate control within guardrails. A communications technology provider, for example, unified its voice, video, and data into standardized workflows, allowing teams to add new capabilities without renegotiating integration on every release. Clear decision rights and a small set of non-negotiables keep risk and reliability in check. The payoff is fewer escalations and faster iteration, allowing teams to ship improvements without constant arbitration.
• Flow instrumentalists use instrumentation and cadence to expose bottlenecks early.  One enterprise software company built its platform around telemetry, automation, and integrated workflows, surfacing quality issues before they escalate. Teams align on shared telemetry and playbooks, concentrating talent where constraints appear. Rather than debating in meetings, teams are empowered to correct the product operations flow, which smooths the process of new products plugging into existing platforms.

Figure 3

Companies deploying these models have identified the subtle risks in the predominant model: forums multiply, decisions bounce upward by default, and cycle time stretches as seam load grows. The updated models aim to reduce those risks.

No one model is right for all situations, and most companies aren’t exclusively one model over the other. In practice, companies deploy various models at different times, depending on strategy, particularly the emphasis between core and adjacency (see Figure 4). Starting point also matters, as the quickest path usually builds on existing strengths.

Figure 4

These evolving models don’t replace seam management. But they do reduce the need to escalate by changing what happens at the working level—how decisions are framed, how authority is delegated, and how work becomes visible before it becomes political.

From theory to action: One concrete move

Tech management teams need to keep evolving their operating models. As leading companies find new ways to move faster and innovate better, others risk falling behind.  With agentic AI becoming standard, models that rely on escalation as the default will feel increasingly slow and brittle at higher speeds.

Start where friction shows up most often, because that’s where trade-offs are clearest. Identify a recurring seam escalation and run a focused pilot to reduce cycle time and escalation while preserving strategic alignment. Anchor the effort in one of three proven approaches, based on your starting point.

• If your organization is meeting-heavy and decisions get relitigated: Pilot stronger decision habits (for example, standard decision memos, explicit trade-off framing, clear pre-read norms, and a “decide once” rule).
• If teams hesitate because decision rights are unclear: Pilot traveling guardrails such as clear delegation, well-defined escalation timeboxes, and transparent red lines.
• If issues surface late and debates rely on opinion: Pilot instrumented flow with end-to-end metrics, shared dashboards that flag anomalies automatically, and a timeboxed shipping cadence.

Define success in both operational terms (cycle time, rework, escalation rate, and throughput stability) and strategic terms (whether local decisions align with strategic intent). Then scale what works. The advantage isn’t choosing the perfect model up front but removing the bottlenecks that slow growth.