Skip to content

The Hidden Algorithm That Decides Which Software AI Will Recommend | Tim Sanders, G2

Eye On A.I.

Hosted by Craig S. Smith

58m
Transcript

Original podcast

Where to listen

Summary

Written by a model from the transcript; not yet reviewed.

This episode explores how artificial intelligence models recommend software and why platforms like G2 have become essential infrastructure for B2B buyers. The guest, Tim Sanders, chief innovation officer at G2, explains that when frontier models like ChatGPT and Google Gemini evaluate commercial intent prompts, they rely on a validation layer rather than simple keyword matching. Because models are trained to avoid regrettable purchases, they heavily weight authoritative third-party trust signals such as online reviews, awards, and accredited lists over vendor claims. The conversation also covers the rise of answer engine optimization, the integration of G2's verified data via model context protocol connectors, and the transition of software-as-a-service companies into AI harnesses.

Listeners get a behind-the-scenes look at how AI search algorithms operate at inference time, how buyer behaviors have shifted toward chat interfaces, and what software vendors must do to remain visible. The discussion provides concrete data points, including research findings on citation click-through rates, the dominance of ChatGPT and Gemini over other models in software research, and the mechanics of avoiding zero-click obscurity in the age of AI search.

  • Software buyers have rapidly shifted toward AI chat interfaces for primary research, with about half of all B2B buyers now starting their search there. 0:15
  • AI models use a validation layer during inference to assess purchase recommendations, checking for trust signals to prevent regrettable buys. 0:38
  • Over forty percent of the algorithm weights driving software recommendations come from an appearance and ranking on authoritative lists. 17:26
  • Software vendors looking to rank in AI search must establish an llms.txt file in their root directory to manage crawling policies. 50:32

Chapters

  1. 0:00
    AI‑Driven Discovery Shift

    The episode opens with a discussion of how B2B buyers now use AI models for software discovery instead of traditional search.

  2. 0:38
    AI Validation Layer and Trust Signals

    The hosts dive into how ChatGPT and Gemini use a validation layer at inference, relying on authoritative lists, awards, and reviews to avoid regrettable purchases.

  3. 1:12
    Tim Sanders’ Background and Vision

    Tim Sanders introduces himself, his Harvard AI Institute role, and his book on high‑touch leadership, framing his perspective on AI trust.

  4. 3:00
    G2’s Mission as the Software Arbiter

    G2 explains its evolution from a crowd‑review platform to a trusted AI‑search partner and its new focus on becoming a trust layer for AI‑driven buying.

  5. 9:00
    Answer Engine Optimization (AEO) and Visibility

    Tim explains the AEO industry, how vendors create content to improve AI visibility, and the difference between being cited and winning the answer.

  6. 15:00
    MCP Connector and AI Data Access

    The conversation covers the MCP protocol, how it lets AI systems retrieve G2’s verified reviews, and its role in automating buyer intent data for vendors.

  7. 22:00
    Future Trends: Agents, Voice, and Guardrails

    The episode concludes with predictions about agent‑based workflows, voice as a breakout category, and the emerging guardrail services needed as AI adoption grows.

  8. 41:59
    G2’s Marketing Services and Review Integrity

    G2’s marketing services are described, emphasizing that they cannot coach reviews and must maintain strict trust and safety to preserve review authenticity.

  9. 49:43
    Voice Reviews and Content Length

    Tim discusses the rise of voice reviews, their longer content, and how they impact AI visibility and user engagement.

  10. 50:29
    Vendor Strategies for AI‑First Buying Journeys

    Tim outlines three key priorities for vendors: making sites machine‑friendly, earning trust signals, and avoiding hard‑to‑crawl formats to win AI recommendations.

  11. 53:08
    Separating Editorial and Advertising

    The hosts discuss the importance of keeping marketing services distinct from editorial content to avoid perception issues.

  12. 53:53
    AI Search and Trust Signals

    G2 highlights how AI search has shifted the need for trust signals, reducing reliance on onsite content and SEO.

Highlights

  • 0:38

    The Validation Layer at Inferenceinsight

    The guest explains that LLMs use a distinct validation process for commercial intent prompts to ensure a recommended purchase will not lead to buyer regret.

  • 5:11

    Integrating via Model Context Protocolstrategy

    The platform utilizes MCP connectors to give AI agents frictionless access to verified reviews and buyer intent data directly within user workflows.

  • 17:26

    Algorithm Weights for Recommendationsdata

    Citing research from First Page Sage, the guest breaks down the specific weights driving software recommendations, showing that 41 percent come from authoritative lists.

  • 22:22

    Comparing Market Share Across Modelsdata

    The discussion reveals that 81 percent of B2B research traffic goes to ChatGPT and Gemini, while Claude is rarely used for primary research despite heavy use in coding.

  • 26:23

    Chatbot Click-Through Rate Realitiesdata

    The guest shares data showing that general citations in AI returns have tiny click-through rates, but citations tied directly to purchase recommendations jump significantly.

  • 40:23

    The Impact of the Entity Updatestory

    The guest describes how OpenAI's August 2025 entity update dramatically shifted citation dynamics, reducing the reliance on raw web scraping in favor of verified signals.

  • 57:36

    The Future of SaaS as Harnessesinsight

    The guest predicts that within five years, a majority of successful SaaS companies will transform from traditional databases into AI harnesses built on skills and context.

Quotes

  • “And you know, Craig, I wrote that with the advent of the internet and digitization, digital transformation.”
    Unnamed speaker · 2:37
  • “Only if they trust it though, and trust is a hard thing.”
    Unnamed speaker · 7:28
  • “All of those are third party. They can't be generated by a vendor. They must be acquired.”
    Unnamed speaker · 17:46
  • “So, so the traffic that you do get is much higher quality. Some might say as high as two, three X.”
    Unnamed speaker · 27:27
  • “as more and more people use AI search to find products, they've got to invest more and more in trust signals and be less and less dependent on, you know, onsite content to win.”
    Unnamed speaker · 53:53
  • “SaaS companies, to quote Satya Nadella, a SaaS company is a cred database with business logic.”
    Unnamed speaker · 57:25
  • “What's a harness? Skills, context, governance, connector tools, those four elements, and that's a harness.”
    Unnamed speaker · 57:47
  • “and their value add is upcharging token spend. They're going to be paid based on consumption.”
    Unnamed speaker · 58:07
  • “Five years from now, seven out of 10 companies today that are successful then will have made the switch from SaaS to harness.”
    Unnamed speaker · 58:22

About this episode

Most companies investing in AI visibility are optimizing for the wrong thing. Being cited by an AI response and being recommended by an AI response are completely different outcomes, with click-through rates that differ by a factor of 70. Tim Sanders, Chief Innovation Officer of G2 and executive fellow at Harvard's AI Institute, joins Craig Smith to explain the hidden mechanics behind how ChatGPT and Gemini actually decide which software to recommend, and why the answer has almost nothing to do with what's on your website. When a user asks a commercial intent question, both models enter a "validation layer" process that specifically down-weights vendor content and seeks verified third-party signals: appearance on authoritative lists (41% of the recommendation weight), awards and accreditation (18%), and online reviews (16%). None of these can be manufactured. They must be earned. The conversation covers the structural transformation in how B2B software buyers behave, half now start their search with an AI prompt, up from 29% a year ago, with two-thirds projected within a year, and why most companies' organic search traffic is on a structural path toward zero. Sanders also delivers some of the most specific competitive data on AI model usage available anywhere: ChatGPT does live retrieval 100% of the time on research queries; Claude does it less than 40% of the time. ChatGPT and Gemini account for 81% of G2's AI research citations. And more than one in four enterprise employees bypass corporate AI tools entirely, using their personal ChatGPT because it has their memory and none of the guardrails. The episode closes with Sanders' most forward-looking prediction: within three years, AI agents will write the prompts, locate the software, and purchase it on behalf of businesses - with minimal human checkpoints - making the trust infrastructure G2 has built over more than a decade the most valuable asset in the AI-driven buying cycle. Subscribe to Eye on A. I. for weekly conversations with the people building and deploying the future of AI.

Transcript

0:00
58:49

10,025 words · 2 speakers

Speaker names marked are inferred from speaking time and the show’s known hosts — they are not voice-verified.

Topics

  • G2
  • Google
  • ChatGPT
  • Reddit
  • AEO
  • OpenAI
  • Gemini
  • Harvard
  • YouTube
  • SEO

Page updated · transcript and summary are machine-generated