The firms gaining ground aren’t buying more AI tools. They’re rewiring how their firms operate. There’s a widening gap between the two approaches — and the firms on the wrong side won’t see it clearly until the economics have already diverged.
At Intapp Amplify 2026, we addressed the question firm leaders are increasingly asking: Is AI structurally embedded in how your firm runs, or is it simply layered on top?
That distinction is not cosmetic. It’s the defining operating model decision of this moment. Over the past year, I’ve seen a clear shift. The firms gaining traction with AI aren’t focused on isolated productivity gains like drafting faster emails or summarizing documents. They’re improving operation leverage by embedding AI across the engagement lifecycle to redesign how growth, compliance, delivery, and revenue interact.
This shift is what separates experimentation from advantage.
For a managing partner, it’s about sustainable growth. For a CFO, it’s about working capital velocity and profitability. For a risk leader, it’s about maintaining control without slowing down engagement teams.
AI will not create marginal efficiency in professional services. It will create structural divergence in profitability and risk posture.
And now, your firm must determine whether AI is strengthening or fragmenting your operating model.
The hidden cost of fragmented AI
Most firms today are in a hybrid state: AI is piloted in business development, automation is applied to parts of compliance, and experimentation is conducted in time capture. Each initiative may move a metric, but none of them move the firm.
Fragmented AI carries a compounding cost that rarely appears on a P&L. Across firms, I’ve seen integration debt rise, governance gaps widen, and change fatigue set in. And the structural leverage that AI should deliver — scaling revenue without proportional cost growth — never materializes.
The firms pulling ahead aren’t layering AI onto their current, outdated workflows. They’re embedding AI directly into the engagement lifecycle and improving profitability as a result.
4 structural shifts already underway
1. Growth is becoming data-enabled
Partner relationships that are stored in inboxes and memory reset when partners leave. On the other hand, firms with embedded relationship intelligence preserve continuity and convert institutional knowledge into wallet-share expansion across priority accounts — even during client leadership transitions.
2. Onboarding speed directly affects revenue timing
Intake delays don’t appear as operational failures. They appear as revenue landing in the wrong month, tightening working capital and making growth more expensive to fund. AI-driven intake and conflicts workflows compress that timeline without reducing oversight. This is not administrative acceleration. It’s balance sheet optimization.
3. Margin leakage is becoming more visible to clients
Clients are increasingly using technology to audit invoices against contractual billing and engagement terms.
When billing compliance is enforced at month end, margin is already at risk. When it’s embedded at the point of entry, realization improves before the invoice is issued.
That improvement compounds across thousands of engagements. Over time, margin protection driven by better compliance often outperforms incremental rate increases.
4. Risk is shifting from episodic to systemic
When professionals pull client data into disconnected AI tools, the boundary between controlled and uncontrolled environments stops being visible. By the time a review happens, the defensibility question is already open.
From my experience, embedding AI within compliance frameworks changes that trajectory. Governance becomes continuous rather than reactive, reducing systemic risk while preserving speed.
This is where many firms underestimate the impact of AI. It’s not simply about accelerating work. It’s about redefining how oversight scales.
The next decade will not be defined by who adopts AI first. It will be defined by whose economics change because of it.
The objection: “We already have systems”
Many firms have CRM platforms, intake tools, and time systems. The question is whether those systems share intelligence or operate in silos.
When growth systems don’t share intelligence with compliance systems — and compliance systems don’t connect seamlessly to billing systems — AI operates in silos. Siloed AI cannot reason across the lifecycle. It can’t surface tradeoffs between opportunity and risk, nor can it optimize operating leverage across the firm.
But agentic AI embedded across a unified data foundation can do all these things — and that distinction is structural rather than cosmetic.
What divergence looks like
The gap will not be visible in a single quarter. It will surface in sustained growth rates, margin resilience, lateral integration success, and client retention. It will appear in audit readiness and competitive positioning.
Firms that embed AI across the engagement lifecycle will:
- Scale revenue without proportional headcount growth
- Shorten cash cycles and improve realization
- Reduce regulatory exposure and strengthen audit defensibility
- Convert institutional knowledge into compounding economic advantage
Early signals indicate that firms that treat AI as an overlay will see complexity rise faster than revenue. The operating model gap will widen until it is irreversible.
AI isn’t just an add-on tool anymore. It’s becoming part of the enterprise backbone, embedded into systems of record, enabling firms to operate at scale.
Uli Hohmann, Corporate Vice President of Cloud and AI at Microsoft, during Intapp Amplify 2026
Join our upcoming webinar, “Growth, redefined: Governed AI across the engagement lifecycle,” where we’ll explore how governed AI can institutionalize growth, improve margin resilience, and reduce systemic risk across accounting and consulting firms.