AI Persuasion Will Arrive Before AI Intelligence – and Sales Tech Knows It

01 /The Concept

Live calls remain an indispensable stage of B2B sales. Whether they happen over video conference or in a conference room, the dynamic hasn't fundamentally changed: a salesperson needs to communicate effectively with a specific human being.

Pepsales built an AI assistant that helps sales reps succeed in those calls – by enabling precise, client-specific personalization of the product demo.

A typical sales conversation unfolds in two movements: probing for the client's problems, then demonstrating how your product solves them. Pepsales provides AI-backed support for every step of both.

Phase 1 – Discovery.

Step 1 is pre-call prep. The AI pulls information about the prospect from the CRM, from recorded previous calls, and from email threads. Based on those inputs, it drafts a meeting plan: a list of questions to ask, a suggested allocation of time for each topic, and a structure for the conversation. The sales rep can review and edit the plan before the call.

Step 2 is the call itself. The AI acts as a real-time prompter in a side panel visible only to the rep. It listens to the conversation, processes it live, and surfaces reminders about questions to ask – along with suggested responses to objections as they emerge.

Step 3 is post-call processing. The AI logs the key takeaways from the call directly into the CRM, generates a list of follow-up actions, and assigns a score to the call. Reps can use this score as personal feedback. The score also feeds into pipeline health estimates for the team.

Phase 2 – Product Demo.

Step 1 is demo preparation. The AI synthesizes the pain points and objections surfaced during discovery, then recommends which specific product features map to each one.

The key insight here: a demo shouldn't tour the full product. It should focus exclusively on the features that address this particular client's particular problems. Clients don't care about your product – they care about an instrument for solving their own goals, which your product might happen to be.

Step 2 is the demo call. The AI again acts as a live prompter, helping the rep follow the pre-planned demo script and suggesting answers to questions as they come up.

Step 3 is post-demo processing. The AI updates the CRM with the outcome, generates a follow-up action list, and scores the demo.

The base plan costs $2,000/month. A free tier with limited functionality is also available, along with an enterprise plan – pricing on request.

Pepsales was founded last year and has now raised its first investment of approximately $1.1 million.

02 /Why It Matters

The underlying shift Pepsales is riding was articulated by Sam Altman: "I expect AI to have superhuman powers of persuasion well before it becomes generally superintelligent."

That's a genuinely powerful insight, and it's spawning a new category of startups.

Uman ([related review](/review/ispolzuj-sverhchelovecheskoe-masterstvo-ubezhdenija)) raised €2.5 million for an AI assistant that helps sales reps build personalized product presentations for each prospect. The AI searches the web and the company's own databases, then extracts relevant materials from past presentations stored on the platform.

AutogenAI ([related review](/review/prostoj-sposob-ubedit)) raised $65.3 million for an AI assistant focused on RFP responses and commercial proposals.

Fluint raised $1.9 million for an AI assistant that generates one-page product summaries tailored to specific stakeholders inside a prospect company, created from the full product deck combined with recordings of prior conversations.

Cuvama raised $4.2 million for a platform that surfaces decision-maker needs and frames them in business case narratives rather than bare feature claims.

Pepsales is built around the demo phase specifically, following the classic principle of "show, don't tell."

But its sharpest feature goes a step further: the AI doesn't just suggest which product capabilities to demonstrate – it also builds the demo environment's test data. And the data in that environment should match what the client actually recognizes.

To show Nike how to handle a delinquent account, your test database needs a delinquent account. To demo the product to a footwear company, the mock product catalog should have sneakers, not cars.

Founders claim that even this seemingly minor detail can increase close rates by 50%.

This also places Pepsales in the growing category of personalized demo creation platforms – a space that's generating its own investment activity.

Consensus ([related review](/review/prodavat-ili-pomogat-pokupat)) built a full-featured platform for automated personalized demo creation and has raised $138.9 million – $110 million of it after that review.

Walnut ([related review](/review/luchshe-odin-raz-uvidet)) raised $56 million for an interactive demo platform, with two additional rounds since – $15 million in 2021 and $35 million in 2022.

Saleo takes a different technical approach: instead of loading custom data on the server, it intercepts what the browser renders and substitutes the right data on the fly. It raised $14.5 million.

What makes Pepsales interesting is that it combines several of these capabilities – discovery coaching, demo preparation, live prompting, and personalized demo data generation – into a single platform at a price point where each of those other tools currently sells as a standalone product.

03 /Opportunities

The broad direction here is clear: build AI platforms that improve B2B sales performance. The specific permutations are varied, as the examples above illustrate.

Among them, Pepsales stands out as the most straightforward entry point – familiar enough in form, but with the right feature set already assembled. That makes it a solid reference architecture for anyone looking to build or compete in this space.