S4E5 – Gemini vs Real Pricing Experts

Summary

Can AI build your pricing strategy? The Pricing Guys give Gemini a common competitive pricing scenario and review its answer point by point.

The Pricing Guys regularly hear the same question: why can’t I just use AI for my pricing strategy?

To find out, they gave Gemini a scenario many software leaders will recognize. A mid-size B2B software company with a loyal customer base and solid margins faces a new AI-native competitor launching at half its price, and the sales team is hearing about it on renewal calls. What should the pricing strategy be for the next 12 months? Gemini returned three recommendations. The hosts found some merit in each, but also found important gaps in context, customer understanding, and execution.

Key takeaways

Gemini correctly warned against matching a 50% price cut, but moved to concessions before establishing whether the threat was real.
The first step is diagnosis: how widespread the competitor mentions are, which customers raise it, and whether any deals have actually been lost.
Value should be defined before any price or concession decision, including security, reliability, and switching costs.
A stripped-down “fighter tier” risks downgrading loyal customers and requires a different go-to-market model.
AI made assumptions not in the prompt, including per-seat pricing and a lack of AI features.
AI is useful for generating ideas quickly, but strategy still depends on asking the right questions and planning execution.

The test: an AI-native competitor at half your price

The prompt was intentionally simple. A mid-size B2B software company with loyal customers and healthy margins learns that a new AI-native competitor has launched at half its price. Sales teams are hearing about it during renewals. Gemini was asked to summarize a 12-month pricing strategy in three key points.

The hosts let Gemini go first, then reviewed each recommendation.

Point 1: Protect renewals through value concessions, not list price cuts

Gemini’s first recommendation was sound at its core: never cut price by 50% to match an unproven competitor. Doing so signals the original price was too high, and margin is hard to recover once given away. Instead, it suggested a “give-to-get” playbook, offering 10% to 15% relief in exchange for multi-year contracts and replacing cash discounts with complimentary enterprise support.

The hosts pushed back on how this plays out with a real customer. If a customer says they are looking at an option that costs 50% less, and the response is a 15% discount tied to a three-year commitment, the customer may feel they were not heard at all. Contract norms also vary widely by market, from annual renewals to 10-year agreements, so any multi-year offer needs to be benchmarked against what customers and the competitor currently sign.

More importantly, the hosts said the strategy started too late in the process. Before any concessions, a company should understand:

How widespread the competitor mentions are, and which customers are raising them
Whether customers are sharing the competitor’s price point or intent to switch
Whether customers see the competitor as equivalent, better, or simply cheaper
Whether any deals have actually been lost

The scenario never said a single deal had been lost.

Start with value, not concessions

A recommendation built on “value concessions” should first define the value. AI-native platforms can be inconsistent, and an apples-to-apples comparison is unlikely. Security, reliability, and the cost of switching all matter. If switching costs are high, the threat may be smaller than it appears.

The hosts recommended reframing the conversation away from price, then segmenting customers, solutions, and where value actually sits. Sales teams also need a prepared response: acknowledge the competitor, explain the trade-offs, and be clear about who it is and is not a good fit for.

They also questioned whether the company needs to respond right away. The next 12 months may be better used as a learning period: track win/loss data, monitor renewal rates and discount depth, and run a few targeted tests. Broad discounting early is hard to reverse, and it teaches the organization to react to every market signal.

Point 2: Create a fighter tier to neutralize price anchoring

Gemini’s second recommendation assumed the competitor is unbundling the platform and selling a lightweight AI feature set. It proposed a good-better-best model within three to six months, with a modular, competitively priced core and a downgrade option for renewal customers who insist on the competitor’s price point.

The hosts noted this contradicts the case for patience. It also creates margin risk. A company with a loyal base is likely selling its premium offering, so introducing a cheaper tier invites existing customers who value the full product to downgrade.

Before building tiers, the company should answer:

Which customer segments should get which package, and why
Where the company wants to win as part of its corporate strategy
What the downgrade risk is and what it means financially, including breakeven
Whether the lower tier is worth pursuing for renewals or new logos

A stripped-down product may also need an entirely different go-to-market approach, such as self-serve sign-up with minimal sales cost and separate branding. That is a significant strategic decision, not a quick pricing fix.

Point 3: Fast-track native AI workflows and shift to a hybrid value metric

Gemini’s third recommendation reframed the threat as a narrative that traditional software is becoming obsolete. It suggested deploying AI features inside existing customer workflows over six to 12 months and moving from per-seat pricing to a hybrid model tied to customer ROI.

The hosts agreed hybrid pricing can align cost with value. The problem was that the prompt never mentioned per-seat pricing or a lack of AI features. Gemini filled in those details itself. Building heavy AI workflows simply because a competitor entered the market, without understanding the business impact, is a risky way to set direction.

What the AI missed: execution and guardrails

Gemini did include execution considerations, which the hosts saw as a strength. But two important pieces were missing.

Communication and internal alignment. Any shift to hybrid pricing or multi-year contracts changes how sales teams sell and how they are paid. Compensation, sales conversations, and internal buy-in all need to be addressed before rollout.

Guardrails and measurement. If a good-better-best model launches and 70% of customers choose the entry tier, is that success or failure? Companies need financial forecasts, breakeven analysis, and clear thresholds that define when a strategy has gone too far or not far enough.

Would you trust AI to build your pricing strategy?

The hosts’ view: AI can produce a strong list of ideas quickly, and compared to a pre-AI world, it can speed up thinking and decision-making. With 20 more prompts, it could likely generate a comprehensive list of options.

But AI answers the questions it is given. Knowing which questions to ask still requires an understanding of the customer, the business context, and whether the threat is real. Strategy comes from connecting those ideas, not executing them one by one.

What companies should learn from the episode

Before responding to a low-priced competitor, a pricing team should be able to answer:

How real is the threat, based on win/loss and renewal data?
Which customer segments is the competitor relevant to, and are they our core segments?
Where does our value differ, including security, reliability, and switching costs?
What is the financial impact of discounts, new tiers, or downgrades?
How will sales compensation, communication, and guardrails support the change?

This is where Revenue Management Labs takes a customized, hands-on approach. AI can speed up analysis and surface options, but it does not replace pricing judgment. The approach needs to reflect the company’s customers, competitive position, data, and specific margin levers.

Recommendations also need to reach the people having renewal conversations every day. A pricing strategy only creates lasting value when teams understand it, can use it, and can measure whether it is working.

The bottom line

Gemini gave a reasonable starting point, and its warning against matching a 50% price cut was correct. But it moved to solutions before diagnosing the problem, made assumptions the scenario did not support, and missed the internal execution work that determines whether a strategy succeeds. AI is a useful tool for building ideas. The judgment to connect them still has to come from people who understand the business.

What this means for your pricing capability

The episode reinforces how RML structures its own work: a pricing capability that lasts is built from three connected pieces, not one quick answer.

Competitive threats test all three. Direction depends on understanding the customer and the threat, tools need to inform decisions rather than make them, and teams need the alignment and incentives to execute.