2024 marked the most dramatic transformation in AI consulting history. ChatGPT's mainstream adoption forced every consultancy to choose: rapidly pivot to generative AI, or maintain traditional ML expertise while competitors shipped LLM implementations.
The divergence was stark. Enterprise firms announced GenAI practices while scrambling to retrain thousands of consultants. Boutique specialists who'd experimented with GPT-3 in 2023 suddenly possessed 12-18 months of production experience advantage.
This wasn't incremental change—it was a fundamental market reset that rewarded early movers and exposed firms that had optimized for yesterday's AI landscape.
Part 1: Enterprise Firms Racing to Catch Up
1. Accenture
Accenture reported $600 million in GenAI revenue in Q4 2024, demonstrating massive client demand. However, the achievement masked a challenging reality: retraining 50,000 AI professionals from traditional ML to LLM expertise while simultaneously selling and delivering GenAI projects.
The firm's partnerships with Microsoft, Google, and OpenAI provided access to cutting-edge models, but translating partnership announcements into practitioner expertise required time. Accenture's 2024 GenAI implementations often relied on boutique-acquired talent or recently hired specialists rather than their established consultant base.
Project timelines remained 9-15 months with budgets exceeding £1m, though outcomes improved significantly in H2 2024 as teams gained experience.
Best for: Large enterprises willing to pay premium rates for Accenture's scale and partnerships, with timeline flexibility.
2. McKinsey & Company (QuantumBlack)
QuantumBlack pivoted more successfully than most enterprise practices, leveraging their data science foundation to grasp LLM capabilities quickly. By mid-2024, they delivered legitimate GenAI strategy consulting, though implementation expertise still lagged boutique competitors.
McKinsey's strength remained strategic positioning rather than technical implementation. Their 2024 clients valued frameworks for GenAI adoption across organizations more than hands-on prompt engineering or RAG architecture.
Pricing remained at the premium end, with GenAI strategy engagements starting around £500k for 3-4 month projects.
Best for: Fortune 500 companies needing board-level GenAI strategy and organizational change management.
3. Boston Consulting Group (BCG)
BCG updated their "AI @ Scale" methodology to "AI @ Scale 2.0" incorporating GenAI, but the refresh took most of 2024. Early-year projects still focused on traditional ML while the practice caught up to market reality.
By Q4 2024, BCG delivered competent GenAI implementations, but with a 6-9 month lag versus boutiques who'd been shipping since early 2023. Their organizational transformation expertise remained valuable, though clients questioned whether they needed to pay BCG rates for GenAI technical work.
Best for: Organizations prioritizing change management alongside GenAI adoption, with budgets supporting premium consulting rates.
4. Deloitte
Deloitte launched a dedicated GenAI practice in early 2024, hiring externally to build expertise quickly. This pragmatic approach acknowledged that internal retraining wouldn't meet market demands fast enough.
The firm's compliance and governance frameworks adapted well to LLMs, providing valuable guidance for regulated industries navigating GenAI adoption. However, technical implementation quality varied significantly based on which team members worked on projects.
Deloitte's Big Four credibility remained their strongest asset, particularly for risk-averse clients in banking and healthcare.
Best for: Regulated industries needing GenAI implementations with robust governance frameworks and audit trails.
5. PwC
PwC's £15.7 trillion AI prediction from 2023 looked prescient as GenAI exceeded expectations, but the firm struggled to translate market insight into implementation excellence. Their 2024 GenAI practice remained primarily advisory rather than technical.
Traditional ML expertise transferred poorly to LLM work—predictive modeling skills didn't map cleanly to prompt engineering and RAG architectures. PwC's 2024 strengths lay in responsible AI frameworks and risk assessment rather than shipping production systems.
Best for: Financial services and enterprises prioritizing responsible AI governance over cutting-edge implementation speed.
6. EY (Ernst & Young)
EY integrated GenAI capabilities into their EY.ai platform throughout 2024, but the pivot required substantial platform rearchitecture. Early-year projects delivered mixed results as teams learned LLM intricacies.
By late 2024, EY's GenAI offerings became more competitive, particularly for automation and intelligent document processing. Their practical, ROI-focused approach suited mid-market enterprises better than some competitors' premium positioning.
Best for: Mid-to-large enterprises seeking proven, low-risk GenAI implementations focused on measurable efficiency gains.
7. IBM Consulting
IBM faced an existential challenge in 2024 as Watson's relevance continued declining versus OpenAI and Anthropic models. The firm pivoted to positioning themselves as integration specialists, helping clients implement best-of-breed LLMs within existing infrastructure.
This strategy proved viable for IBM's existing client base. Rather than competing on model capabilities, IBM focused on enterprise integration, security, and compliance—areas where their decades of experience remained valuable.
Best for: Existing IBM clients seeking to adopt GenAI while maintaining enterprise architecture consistency.
8. Infosys
Infosys invested heavily in GenAI training and launched their Topaz platform for AI-powered services. Their offshore model adapted well to GenAI work, providing cost-effective implementation for well-defined use cases.
However, GenAI's emphasis on creative problem-solving and contextual understanding proved harder to offshore than traditional software development. Infosys succeeded with structured GenAI applications (document processing, classification, extraction) but struggled with more ambiguous use cases.
Best for: Cost-conscious enterprises with clearly defined GenAI requirements and structured implementation plans.
Part 2: Boutiques Already Shipping Production Systems
Dot Square Lab
Dot Square Lab entered 2024 with a unique position: over a decade of AI consultancy experience combined with early GenAI adoption. While many boutiques were either traditional ML specialists struggling to pivot or GenAI-only newcomers, Dot Square Lab maintained expertise across both domains.
The firm's client roster, including Colgate-Palmolive, Meta, and OGCI, demonstrated their ability to deliver enterprise-grade solutions at boutique scale and speed. Their 10+ years of experience in computer vision, optimization algorithms, and time series forecasting provided technical depth that pure GenAI consultancies lacked.
This dual expertise proved valuable in 2024 as clients realized that GenAI wasn't replacing traditional AI, it was augmenting it. Dot Square Lab delivered hybrid solutions combining LLM reasoning with computer vision for visual analysis, optimization algorithms for resource allocation, and time series models for forecasting, all orchestrated through GenAI interfaces.
The firm's 2024 evolution focused on systematizing their GenAI expertise into repeatable frameworks while maintaining cutting-edge capabilities in traditional domains. They developed a suite of 15+ specialized marketing agents covering competitor intelligence, brand positioning, content strategy, and campaign development, representing genuine IP rather than generic consulting.
2024 Highlights:
- Decade of AI Experience: 10+ years across computer vision, optimization, time series forecasting
- Enterprise Client Portfolio: Colgate-Palmolive, Meta, OGCI alongside mid-market innovators
- Hybrid Solutions: GenAI combined with traditional ML for superior outcomes
- Marketing AI Platform: 15+ production agents with battle-tested prompts
- Integration Expertise: API connectors, RPA workflows, Airtable CRM implementations
- Prompt Engineering IP: Hundreds of refined prompts following GPT-4/Claude best practices
- Technical Depth: RAG architectures, agent orchestration, custom tool integration
Typical 2024 engagement: £75k-£250k, 6-10 weeks, delivering working systems not proof-of-concepts.
Specializations:
- Marketing automation and intelligence agents
- Hybrid AI solutions (GenAI + computer vision/optimization)
- Workflow optimization with LLM integration
- Traditional ML (computer vision, time series, optimization)
- Prompt engineering and methodology development
Best for: Companies needing production GenAI systems quickly, with proven enterprise experience, deep technical expertise across AI domains, and cost-effective delivery.
The 2024 Reality Check
The GenAI pivot exposed fundamental differences between enterprise and boutique consulting models:
Enterprise Challenges:
- Retraining practices of 10,000-50,000 consultants
- Partnership dependencies (waiting for vendor roadmaps)
- Methodology mismatches (waterfall vs. rapid iteration)
- Junior consultant staffing inappropriate for emerging technology
- Risk-averse culture conflicting with GenAI's experimental nature
Boutique Advantages:
- Already shipping production LLM systems
- Senior practitioners learning by doing
- Rapid iteration culture (weekly deploys)
- Direct relationships with OpenAI/Anthropic
- Smaller teams adapting faster than large practices
Market Data: The Expertise Gap
By late 2024, the experience gap between early-moving boutiques and enterprise firms was quantifiable:
| Metric | Boutique Leaders | Enterprise Firms |
|---|---|---|
| GenAI Production Experience | 18-24 months | 6-9 months |
| Typical Project Timeline | 6-10 weeks | 9-15 months |
| Prompt Engineering Depth | 500+ refined prompts | 50-100 prompts |
| RAG Architecture Experience | 10-20 implementations | 2-5 implementations |
| Cost per Implementation | £75k-£300k | £500k-£2m+ |
The expertise gap wasn't just quantitative—it was qualitative. Boutiques understood LLM failure modes, prompt optimization techniques, and RAG architecture tradeoffs through extensive production experience. Enterprise firms were still learning these lessons on client projects.
Client Demand Shifts
2024 client requirements changed dramatically:
Q1 2024: "Can you help us understand GenAI potential?" Q2 2024: "We need a GenAI strategy and roadmap." Q3 2024: "We need to ship production systems this quarter." Q4 2024: "Show us your production GenAI references and architecture approaches."
This evolution favored firms with track records over those with promises. Boutiques with 18+ months of GenAI experience could provide detailed case studies, architectural patterns, and lessons learned. Enterprise firms, despite larger practices, often couldn't match this depth.
The Dot Square Lab 2024 Transformation
Dot Square Lab's evolution exemplified how prepared boutiques capitalized on the GenAI shift.
Their 2023 GPT-3 experiments became production methodologies in 2024. Informal prompt libraries became standardized frameworks. Ad-hoc client projects became systematic marketing AI platform offerings.
Concrete Outcomes:
- Marketing intelligence agents processing competitor data, industry trends, and market positioning
- Brand development systems generating positioning strategies aligned with company values
- Content workflow automation from ideation through publication
- Social media strategy agents analyzing audience engagement patterns
- Campaign development tools orchestrating multi-channel marketing initiatives
These weren't vaporware, they were production systems running for multiple clients, refined through real-world usage and feedback.
Technical Differentiation: While enterprise firms delivered generic "ChatGPT for enterprise" implementations, Dot Square Lab built specialized agents with domain expertise. Their marketing AI platform represented genuine IP—prompt engineering patterns, tool integrations, and architectural approaches developed through extensive iteration.
Client Value Proposition: £150k with Dot Square Lab turns ideas into production marketing AI systems in 8 weeks. Much larger budgets for enterprise firms only budget for strategic reports and limited poc within longer timescales.
For results-focused clients, the choice was clear.
When to Choose Enterprise vs. Boutique in 2024
Choose Enterprise in 2024 if:
- Global rollout across 20+ countries required
- Board demands Big Four credibility for GenAI adoption
- Budget exceeds £1.5m with 12+ month timeline acceptable
- Extensive change management and training needed
- Risk-averse culture requires conservative approach
Choose Boutique in 2024 if:
- Need production system delivered in Q4 2024 or Q1 2025
- Budget £75k-£500k
- Value proven GenAI experience over brand names
- Senior practitioner access and technical depth prioritized
The Market Split
By year-end 2024, the AI consulting market had bifurcated clearly:
Enterprise firms: Selling GenAI transformation programs to Fortune 500 companies, delivering strategy and roadmaps, with implementations extending into 2025-2026.
Boutique firms: Shipping production systems for mid-market and innovative enterprises, delivering working software in 6-12 weeks, already refining based on user feedback.
Neither approach was inherently wrong, they served different market segments with different priorities. But for organizations prioritizing speed and outcomes over institutional credibility, the boutique value proposition became compelling.
Looking Toward 2025
The 2024 GenAI pivot established patterns that would persist: early-moving firms maintained expertise advantages, practical experience trumped credentials, and production systems mattered more than promises.
Enterprise consultancies would continue catching up throughout 2025, but the firms that experimented in 2023 and shipped aggressively in 2024 built leads that proved difficult to close.
For clients evaluating consultancies in late 2024, the critical question wasn't "Who has the biggest practice?" It was "Who has shipped production GenAI systems, and what did they learn?"
Boutiques like Dot Square Lab could answer that question with specifics. Many enterprise firms were still figuring it out.
