Forget Playbooks: Selling AI Is About Strategy, Not Scripts

by Mayur Potdar

‍

After 15+ years in sales, marketing, and business development — and having spent the last few years working with an AI startup — I can tell you with absolute certainty: selling AI is not like selling SaaS, tech products, or traditional enterprise solutions.

Walking into AI sales with a SaaS playbook is a great way to hit a wall.

Why? Because selling AI isn’t about showcasing features or talking speeds and feeds. It’s about solving real, complex problems in ways that require consultative thinking, deep listening, and careful alignment between tech capability and business need.

Let’s break down what sets AI sales apart — and what I’ve learned the hard way.

‍

1. There’s No Script for Selling AI — It’s 100% Consultative

‍

In SaaS, you often have a defined category: CRM, ERP, HRMS, etc. Buyers know what to expect. You show your features, demo the UI, and compare pricing.

AI? Different game.

There is no fixed playbook. Each conversation can — and often does — go in a different direction. You can’t lead with a templated pitch. Instead, you’re helping the prospect discover what AI could do for them. That means you have to:

• Understand their business pain first
• Explore if AI is the right approach
• Co-create use cases together

It’s more like being a solution consultant or a transformation advisor than a product seller. You’re not just selling a product — you’re helping them reimagine a process or workflow.

👉 Takeaway: Your sales team should think more like AI solution architects than quota-crushing SDRs.

‍

2. ICP Is Everything: You Can’t “Spray and Pray” in AI

‍

Early in my AI journey, we tried the typical SaaS tactic — bulk outreach, generic messaging, wide top-of-funnel. It bombed. AI isn’t one-size-fits-all. It’s highly contextual, and use cases vary dramatically across industries.

• A logistics company wants route optimization.
• A bank cares about fraud detection.
• A retailer wants demand forecasting or hyper-personalization.

Each of these needs completely different data, teams, metrics, and stakeholders. That’s why choosing a narrow, high-propensity ICP (Ideal Customer Profile) is non-negotiable. You need to focus on:

• Industries where AI is already being explored
• Functions with clear pain points (e.g., compliance, operations, support)
• People who control budgets and feel the problem deeply

Trying to pitch AI to 1,000 random companies dilutes your credibility. Instead, go deep on 50 well-researched accounts where you know there’s a fit.

👉 Takeaway: Narrow down your ICP to high-propensity, problem-aware, budget-owning buyers in 1–2 specific verticals. Then go deep.

‍

3. Pitch Outcomes, Not Algorithms

‍

AI is attractive on paper. LLMs, generative insights, embeddings — it all sounds impressive. But most business buyers don’t care about the tech. They care about what it can do for them.

What I learned early on is: stop pitching capabilities — start mapping them to real business outcomes.

Instead of:

“Our AI can analyze unstructured text.”

Say:

“We help banks reduce manual compliance review time by 40% using AI summarization of regulatory reports.”

Buyers want results, not models. They don’t want a GPT-4-powered system — they want to:

• Reduce costs
• Speed up operations
• Improve customer experiences
• Minimize risk

And to earn trust, you need to be brutally honest about what’s possible and what’s not. Overpromising is the fastest way to kill deals (and damage reputation).

👉 Takeaway: Anchor your pitch on business outcomes, not technical features. And don’t promise what your model or team can’t reliably deliver.

‍

4. Choosing the Right Market for Zero-to-One Growth: India vs. US Challenges

‍

If you’re taking an AI product from 0 to 1, market selection is everything. Especially when you’re bootstrapped or in early stages. We initially tried selling in India — home turf, right? But we hit a wall. Here’s why:

• Budgets for AI are limited, especially in mid-market companies
• Long decision cycles and lower experimentation appetite
• Often, a “wait and watch” approach unless there’s proven ROI

When we started targeting the US, we saw a major shift:

• Buyers were more open to innovation
• Clearer budgets for “AI and automation”
• More maturity in data infrastructure and digital transformation

That’s not to say India isn’t a good AI market — it is, for certain verticals like BFSI or government. But for early-stage growth, selling to US or global enterprises with mature pain points and innovation budgets can be far more fruitful.

👉 Takeaway: Choose your first 5 customers wisely. Their feedback and success stories will shape your GTM flywheel.

‍

5. Cultural and Buying Method Differences: India vs. US

‍

Another key difference I noticed: the approach to buying and risk.

• US buyers are often forward-thinking, pilot-friendly, and ready to test new ideas if they see strategic value. You can co-create use cases with them.
• Indian buyers, especially in large enterprises, tend to be more cautious, hierarchical, and focused on proven solutions. Innovation decisions can get buried under layers of approvals.

Also, Indian clients often expect:

• More handholding
• Faster support
• Pricing flexibility

US clients, on the other hand, are okay with structured onboarding, self-service, and SaaS-style models — as long as the ROI is clear and measurable. Understanding these cultural nuances changes everything: your pricing, support model, onboarding, and even sales scripts.

👉 Takeaway: In India, build credibility before you build revenue. In the US, drive fast toward ROI metrics and integration timelines.

‍

Additional Insights

‍

Here are a few more key differences that are often overlooked:

• Proof of Value > Demos
  SaaS sales can often close on a good demo. AI deals usually need a PoC or pilot to prove the value in the customer’s context. Paid PoC or pilot is a good validation of the seriousness of the client.
• Cross-Functional Stakeholders
  AI buying rarely involves a single buyer. You’ll need to align product, data science, business ops, compliance, and IT — all with different agendas.
• Data Access Is the Real Bottleneck
  Even if you have the perfect AI solution, most customers aren’t ready with the data they need. Anticipate delays and integration challenges.
• Evangelism Matters
  You’re not just selling a product — you’re educating a market. Thought leadership, webinars, use-case storytelling, and industry-specific content are essential.

‍

Conclusion

‍

Selling AI products is fundamentally different from traditional SaaS or tech sales. It requires a consultative approach, razor-sharp ICP selection, deep alignment of product capabilities with industry pain, and a nuanced understanding of market and cultural dynamics.

For those willing to invest the time and thoughtfulness, AI sales offer an exciting opportunity to transform businesses and reap rewards in a rapidly growing market.

#AISales #TechSales #SaaS #AIProduct #ConsultativeSelling #GoToMarket #IdealCustomerProfile #MarketStrategy #SalesLeadership #B2BSales #IndiaTech #USMarket #ArtificialIntelligence #BusinessDevelopment #SalesStrategy #DigitalTransformation

‍

You can also check this post on Medium.

‍

‍