The AI services test

For those building, investing in, or preparing to exit a data & AI consultancy, the key question is: which consultancies will be able to capture demand consistently and create durable value from it. Across the market, five characteristics distinguish the firms best placed to capture value.

1. Focus on AI foundations before AI use cases

AI is only as effective as the data infrastructure beneath it. Many organizations are experimenting with generative and agentic AI, but fragmented data estates, inconsistent governance and legacy architecture often prevent those initiatives from scaling and driving measurable value.

Consultancies focused primarily on downstream AI use cases may capture near-term spend, but they risk being deprioritized when the AI fails to deliver value because the foundations were not in place.

In contrast, consultancies focused on defining AI strategy and building scalable data foundations will find themselves better positioned as long term strategic partners for their clients, with AI implementation soon becoming a natural extension of their scope.

2. Real technical depth

Data & AI consultancies operate inside vendor ecosystems. While some have their own IP, most of the demand comes from configuring, implementing and managing third-party data platforms and tools. Vendor alignment is therefore a major driver of credibility, differentiation, and pipeline.

The strongest consultancies usually combine one or two anchor relationships with a small number of complementary partnerships. This reflects how enterprise typically buy: data stacks are fragmented and modular, so consultancies with capabilities in supporting tools alongside the core data platform can provide more holistic, and therefore more valuable support.

The question is then: which vendors to anchor to? Given the pace of innovation, the optimal choice is constantly shifting and depends on the target ICP, as each data platform offers different strengths. Hyperscaler offerings such as Microsoft Fabric and Amazon Redshift are well suited to smaller and mid-market organizations, and particularly in the case of Microsoft, provide access to a structured partner programme, co-sell infrastructure and referral engines that IT services investors have long valued. As the market matures, sustained differentiation may increasingly require expertise in specific layers of the vendor stack, for example Fabric for data engineering or Purview for governance, with clear proof points and repeatable delivery.

Outside the hyperscalers, enterprise vendors like Databricks or Snowflake are in rapid growth and increasingly investing in their partner ecosystems. Earlier specialization in these platforms can generate outsized returns through vendor-led referrals and reputation advantages before the market becomes overly crowded.

3. Owns IP that changes the economics

AI is increasingly improving delivery efficiency. As automation reduces manual effort required for coding, testing and documenting, the link between time input and value delivered weakens. This creates both risk and opportunity. Consultancies that rely heavily on manual, repeatable delivery may face pricing pressure as clients expect productivity gains to be shared. Consultancies that embed repeatable IP into their delivery model can turn that shift into operating leverage. Proprietary accelerators, tools and reusable frameworks automate common tasks, shorten timelines, and improve consistency. They can also support sales motions by demonstrating innovation and faster time to value. Done well, this supports margin and reduces reliance on linear headcount growth.

In some cases, IP can also be productized into repeatable propositions, managed services or platform-like offerings. This can create more recurring revenue, deepen client relationships and provide a higher-margin growth lever. The test is whether the IP changes how the consultancy scales; if it does not, it is more likely to be sales collateral than a sustainable source of differentiation.

4. Embedded at a strategic level

The most durable demand comes from large enterprises. Their data estates are complex, business-critical, and constantly changing, which sustains multi-year external support rather than one-off delivery. This is especially true in regulated and data-rich sectors, where investment in data and AI is more resilient than in margin-constrained markets.

The strongest consultancies serving this segment are those embedded at a strategic level, rather than just providing delivery capacity. Firms that advise on data strategy, target architecture and governance work closer to decision makers and across functions, and have greater visibility of opportunities for follow-on work.

Where a firm sits in an organization also shapes the relationship with global systems integrators, which capture a large share of spend in this market. Specialist consultancies sitting at a strategic level do not always need to displace systems integrator (SI) spend – instead they can sit above or alongside them, providing niche expertise and agility while the SI provides scale. The key distinction is between being strategically embedded versus being brought in for one-off project work. The former supports repeat demand and pricing power, the latter is more exposed to lumpy revenue and commoditisation.

5. Wins the talent battle

There is a structural shortage of senior data and AI talent. Enterprises often struggle to recruit and retain the senior technical specialists required to build modern data platforms and production-grade AI solutions. That creates demand for external partners, but it also makes the consulting model difficult to scale.

The most attractive consultancies have an effective recruitment engine for senior talent, and can offer the complex project exposure, strong culture and progression opportunities to retain them.

The structure of the pyramid also affects resilience to AI disruption. Consultancies built around junior implementation capacity are more exposed as AI automates lower-level delivery tasks. Senior-heavy firms with strategic, architectural and governance-led capabilities are likely better insulated. In this segment, talent depth is a core driver of both delivery credibility and valuation.

Testing for durability

AI demand is strong, but durable value in data & AI consultancies comes from structural positioning rather than momentum alone.

When assessing a consultancy operating in this space, five questions matter:

• Is it anchored in data engineering and governance, or primarily selling downstream AI use cases?
• Does it demonstrate genuine technical depth within its vendor ecosystem?
• Does it own IP that supports repeatability and scalable delivery?
• Is it embedded strategically within large, data-intensive enterprises?
• Can it consistently attract and retain senior talent?

Consultancies that answer these convincingly are better positioned to convert AI demand into sustainable growth and defend valuation over time.

We continue to track developments in the data & AI consultancy market. If you would like to discuss what we are seeing in the sector, we would be glad to continue the conversation