A New Way to Look at Enterprise Technology in 2027+

Get my new posts in a weekly newsletter. Subscribe.

It may seem early to revisit the enterprise technology framework for 2026 in August, but considering the dramatic compression of cycles, and the fact that the first eight months provide enough evidence to test the assumptions we carried into this year, its time to update the framework used to discuss and plan.

Earlier this year, the planning work centered on AI transformation, resource optimization, infrastructure maturation, and trust and governance at scale. AI transformation is progressing more slowly and unevenly than anticipated. Infrastructure became broader than cloud. Governance expanded into identity, machine authority, provenance, accountability, and continuous control. Resource constraints became more visible as AI moved into operating environments. The larger problem is that the way we look at enterprise technology did not age as well.

The old taxonomy was organized primarily around technologies: cloud infrastructure, hybrid cloud, cybersecurity, compliance, generative AI, agentic AI, edge computing, quantum computing, and related areas. That was useful when the principal uncertainty was whether these technologies were mature enough to matter and whether enterprises should adopt them. Those decisions have moved forward. Enterprises are already adopting AI, cloud, automation, and other technologies while continuing to operate decades of accumulated applications, infrastructure, data, processes, controls, and vendor relationships.

AI provides the clearest evidence of the change. PwC returned this summer to 351 CEOs it had surveyed in late 2025 and found that the aggregate proportion reporting AI-related revenue gains, cost reductions, or both remained around one-third despite another eight months of activity. BCG found that nearly nine in ten CEOs could identify some AI benefit in targeted areas, yet only 26% had embedded AI within broader transformation, and only 14% clearly defined P&L impact across all AI initiatives. Adoption continues, while enterprise impact progresses at a different pace.

The updated framework starts with a simple premise: the taxonomy describes the enterprise; technologies map onto it. That gives it a longer useful life. AI can accelerate, slow, fragment into new approaches, or eventually become routine without forcing another redesign. The enterprise will still need to create value, organize people, preserve what it knows, maintain an adaptable technology estate, and retain control.

I am changing the way i look at the enterprise technology conversation around five categories.

• Business Value examines whether technology produces measurable improvement in enterprise performance and whether those improvements justify continued investment and organizational change. Business value is broader than ROI or direct financial return. Revenue, margin, and cost matter, but so do customer outcomes, decision quality, cycle time, risk reduction, workforce effectiveness, resilience, and new business or operating models. Technology adoption and productivity establish activity; business value requires observable consequences.
• Workforce & Organization examines how technology changes roles, organizational structures, human and machine responsibilities, expertise requirements, and management. AI agents make this visible, but the issue is broader than AI. Enterprises must decide which responsibilities remain human, which can move to machines, how people supervise automated work, and how to preserve experienced judgment and institutional knowledge as roles change.
• Enterprise Intelligence & Context expands the traditional data discussion. Useful enterprise decisions depend on records and data, but also on semantics, business rules, operating procedures, institutional memory, decision histories, relationships, and proprietary knowledge. Models, applications, and vendors will continue changing. The strategic issue is whether the enterprise retains sufficient control over what it knows and can continue using that knowledge regardless of which technology consumes it.
• Architecture, Infrastructure & Modernization addresses the technology estate as a connected system. Compute, storage, memory, networks, power, cloud, applications, SaaS, integrations, legacy systems, and technical debt increasingly have to be considered together. Architecture determines how easily the enterprise can add new capabilities, move workloads, change providers, modernize applications, and retire obsolete technology. Modernization must include removal as well as addition; otherwise, each technology cycle becomes another permanent layer.
• Trust, Identity & Control covers the mechanisms that govern technology behavior. Trust as an enterprise outcome is the confidence stakeholders place in the organization over time. This category is different: it includes cybersecurity, identity, privacy, compliance, machine authority, provenance, auditability, and technology dependency. As software increasingly acts rather than informs, enterprises need to establish who or what acted, what authority was granted, what information was used, what happened, and who remains accountable.

These five categories are evaluated through three enterprise outcomes—Growth, Resilience, and Trust—and two constraints—Enterprise Capacity and Talent & Judgment. Capacity recognizes that capital, compute, infrastructure, management attention, time, energy, and organizational ability to absorb change are finite. Talent & Judgment recognizes that expertise can often be acquired faster than operating experience and sound judgment can be developed.

Technologies then become overlays. An agentic AI initiative may affect Business Value through process redesign, Workforce & Organization through role changes, Enterprise Intelligence & Context through access to proprietary knowledge, Architecture through integration and compute requirements, and Trust through identity and delegated authority. The same approach applies to cloud, robotics, automation, edge, or whatever technology becomes important next.

Customer, industry, function, and operating domain work the same way. Apply them after the enterprise problem is classified. The relevant customer changes by company, business model, geography, product, channel, and use case. Embedding those variables into the taxonomy would eventually create hundreds of versions of the same framework. The taxonomy provides the stable structure; customer and operating context are added when it is applied.

Frontier technologies remain outside this taxonomy because the planning horizon is different. This framework is designed for enterprise decisions within the next 24–36 months. Frontier research looks further ahead and evaluates readiness, triggers, leading indicators, potential value, and time to scale. When a frontier technology moves into the active planning horizon, you can map its consequences into the same five categories.

The taxonomy is also not intended to produce strategy by itself. Its role is classification and allocation: identify the enterprise problem, determine where it belongs, connect it to the relevant outcomes and constraints, and expose dependencies across categories. Strategy begins after that work, when priorities, trade-offs, sequencing, investment choices, ownership, and governance must be resolved through the appropriate frameworks and experienced judgment.

Over the next five newsletters, I will take each category separately and test it against what we learned during 2026. The objective is to create a more durable starting point for enterprise technology strategy in 2027: one based on what technology is doing to the enterprise rather than whichever technology happens to dominate the market conversation. And, just in time for planning and budgeting.

Here are some reading resources: