Generative AI has changed the economics of creating a presentation.
With Claude, ChatGPT, Microsoft Copilot, Gemini and a growing range of AI-powered presentation platforms, a consultant can develop a storyline, summarize research, draft slide content, generate visuals, and increasingly create or refine the deck itself. Work that used to take hours can take a fraction of the time.
The cost of AI can be low compared with the hourly cost of presentation production. For consulting firms and other professional services organizations, though, the bigger cost is the time of whoever is using the AI—and what else that person could be accomplishing with it.
The AI Cost Is Only Part of the Equation
When a consultant uses AI to build a client deck, the first draft is accelerated. After that, the consultant may still have to write and refine prompts, check the output, restructure the storyline, fix formatting and layouts, find or make visuals, apply firm and client branding, verify data and citations, work in stakeholder revisions, and do a final quality check.
AI can reduce the time required for many of these steps, but it doesn’t eliminate them. The greater opportunity for savings comes from reducing the amount of high-cost professional time spent on production, refinement, and quality assurance.
The relevant calculation therefore isn’t AI cost vs. presentation production cost.
It is: AI + professional time + iteration + design + quality assurance + rework vs. specialist production cost + limited professional review time.
The Most Expensive Part of an AI-Generated Deck May Be the Person Creating It
Consider a hypothetical team preparing a 30-slide client presentation. Suppose generative AI reduces the presentation-production time from three hours to one hour, giving two hours back to the consultant.
Now consider a different setup. The consultant spends 15 minutes briefing a specialist presentation team and another 15 minutes reviewing the finished product. The specialists handle production, formatting, visual design, brand compliance and quality assurance, potentially using AI within their workflow to increase speed and efficiency.
The consultant’s involvement drops from one hour to 30 minutes.
The firm’s decision then comes down to whether those 30 minutes of consultant time are worth more than the added cost of specialist production. Across thousands of presentations and hundreds or thousands of well-paid professionals, seemingly small differences in production time can become material.
Professional services firms sell expertise and judgment. Every hour a consultant spends formatting slides, adjusting charts, searching for images, or correcting an AI-generated presentation is an hour that cannot be spent analyzing a client’s problem, developing recommendations, building relationships or pursuing new business.
AI Doesn’t Eliminate the Need for Judgment
AI can deliver significant gains in speed and quality, but its capabilities remain uneven. A field experiment involving 758 strategy consultants found that, for tasks inside AI’s capability boundary, professionals using AI completed work 25.1% faster and produced results rated more than 40% higher in quality. For a task outside that boundary, however, AI users were 19 percentage points less likely to reach the correct answer.
That distinction matters. Professionals should use AI where it improves their judgment and speed, but access to a tool does not remove the need to decide which tasks belong with the professional, which belong with the technology, and which belong with a specialist.
The same applies to presentation quality. AI can get a deck much closer to completion, but someone still needs to recognize when the output is generic, when a presentation that is 90% complete is not client-ready, and when it is faster to stop prompting and solve the problem directly.
Three Models, Not Two
The choice shouldn’t be framed as AI versus outsourcing. There are at least three models firms should evaluate, each capturing a different share of AI’s time savings.
1. Professional + AI Self-Service
The professional uses AI tools directly to build and refine the deck.
This can work particularly well for early-stage thinking, internal presentations, simple decks, outlines and situations where speed matters more than production quality.
The professional can move quickly from idea to output, but an expensive resource is still doing the production.
2. Specialist Presentation Support
Presentation production is routed to a dedicated internal or external team.
This model can be particularly effective for high-volume or client-facing presentation work where consistency and production quality matter.
3. AI-Enabled Managed Presentation Services
The third model combines the advantages of both.
AI is embedded into the presentation-production workflow, while specialists provide the design judgment, brand knowledge, quality assurance and human oversight required for client deliverables.
AI makes the managed service more productive, while the consultant isn’t being asked to become an expert AI presentation operator.
A Better Business Case for Presentation AI
When evaluating these models, firms should measure the full economics of presentation production.
A useful framework should include:
Cost/Value Factor:
AI/software cost
Professional + AI:
Low
Specialist Support:
Varies
AI-Enabled Managed Service:
Embedded/varies
Cost/Value Factor:
Professional time
Professional + AI:
Medium
Specialist Support:
Low
AI-Enabled Managed Service:
Low
Cost/Value Factor:
Specialist production
Professional + AI:
Low
Specialist Support:
High
AI-Enabled Managed Service:
High
Cost/Value Factor:
Brand/design expertise
Professional + AI:
Variable
Specialist Support:
High
AI-Enabled Managed Service:
High
Cost/Value Factor:
Quality assurance
Professional + AI:
Professional-led
Specialist Support:
Embedded
AI-Enabled Managed Service:
Embedded
Cost/Value Factor:
Ability to scale
Professional + AI:
Medium
Specialist Support:
High
AI-Enabled Managed Service:
High
Cost/Value Factor:
Opportunity cost of professional time
Professional + AI:
Medium
Specialist Support:
Low
AI-Enabled Managed Service:
Low
The weighting will differ by firm and by type of presentation. There is unlikely to be a single model that makes sense for every deck.
Route the Work, Not Just the Technology
A five-slide internal presentation is an obvious candidate for AI self-service. A major pitch, a board presentation or an 80-slide deliverable may benefit from an environment where AI accelerates the work, but experienced presentation specialists remain responsible for design, brand and quality. High-volume organizations may need both.
The routing decision should be based on value, complexity, risk, volume, turnaround, and the level of brand or design judgment required—not simply on whether the technology can generate a slide.
As AI tools become more capable, this distinction will become even more important. The question is not simply how much work AI can perform, but how firms combine technology, professional expertise and specialist resources to create the most efficient workflow.
Don’t Use AI to Make Expensive People Better at Low-Value Work
Making a consultant twice as fast at formatting a presentation is useful. Taking most of the formatting work off their plate altogether may be considerably more valuable.
For professional services firms, the next phase of AI adoption should be about more than giving professionals powerful tools. Instead, use them where they genuinely improve the workflow, and preserve professional time for judgment, relationships and expertise where it creates the greatest value.
Keeping AI costs down matters less than maximizing the value of human time.
