The Discipline After the Experimentation ft. Simon Farthing (Bloomreach)

The barrier to building has never been lower. The question of who owns what gets built has never been harder to answer.

Simon Farthing, VP of Customer Success and Client Services at Bloomreach, joined host Josh Schachter for the [Un]Churned episode “204. He Cut Exec Prep From 4 Hours to 10 Minutes.” They cover what 60 tool evaluations across 18 months actually taught his team, how he built a personal mission control that turned four hours of exec prep into ten minutes, and what institutional AI leadership looks like inside a 100-person post-sales org.

---

Listen on YouTube, Spotify, Apple Podcasts, and Gainsight.com.

Subscribe now and get the [Un]Churned take straight to your inbox.

---

🎯 The [Un]Churned Take: Who Owns the Outcome?

The AI outputs filling our workflows right now are good enough to be dangerous. So good, you might find yourself skipping that second check or easing up on your LLM interrogations. In most cases, good enough gets the job done. Until it doesn’t.

The question of accountability isn’t new, but AI makes it harder to answer. When a Customer Success Manager (CSM) prepares a QBR by hand, it’s clear who is responsible. But if an AI tool pulls data from three sources, another model formats it, and someone reviews it for ten minutes before a customer call, who is accountable? Is it the CSM who reviewed it, the tool that made it, or the team that picked the tool?

Most of us are figuring out as we go, but if you never tackle the question directly, you risk never addressing it at all.

At the Individual Level, You’re the Stopgap

On this week’s episode of the [Un]Churned podcast, Simon Farthing, VP of Customer Success at Bloomreach, gives us a good place to start. A self-labeled “failed computer science graduate,” he had the basic knowledge to build software, but the power of vibe coding gave him the tools he needed to bring his ideas to life. He created a personal mission control in Claude that brings together customer health data, one-on-one tracking, and executive prep into one place. What used to take three or four hours now takes him just ten minutes.

“I now get that in 10 minutes, and then I can spend the rest of the time really interrogating the insight, really building the talk track, and hopefully it’s making me better at my job and more useful to the customers who I speak to because I know more about their businesses than I would have done in the past.”

The mission control works because Simon’s the one responsible. He’s the only user and checks every output before it’s used. If something’s wrong, he finds it because he handles both creating and reviewing the work. His standard for his broader team reflects the same logic:

“The sentiment we have in my team is anything that you produce, ‘AI did it’ is not an excuse.”

On an individual level, responsibilities are cut and dry. If you build it and review it, you own it. When one person is involved, accountability is easy to map back to.

When the Workflow Gets More Complex, So Does the Answer

As AI continues to rapidly accelerate, it’s taking on more complex workflows. For example, the renewal process. Renewals aren’t handled by one person. They involve CSMs, account managers, finance, legal, and executive sponsors. Data comes from a variety of systems, and the process needs consistent results. With retention more important than ever, AI now plays a role at every stage, whether that’s gathering health scores, flagging risks, or generating renewal decks.

When something goes wrong in a renewal—like a wrong number, a missed risk, or a customer leaving despite a good health score—it’s hard to know who’s responsible. Was it the data, the model, the CSM who didn’t look closer, or the leader who set up the process?

This is the layer where the industry is starting to ask a different kind of question. In a recent article in the Silicon Valley Business Journal, Gainsight CEO Chuck Ganapathi described how software companies are rethinking what they actually sell as AI changes the equation. Rather than simply providing tools for customers to operate, some vendors are moving toward owning the work itself. As Chuck put it, “The vendor should own the outcome.“

In other words, the role of accountability is moving. If a vendor manages renewals with AI and human oversight, they’re not just giving customers a tool; they’re responsible for the outcome. Now, accountability shifts from the customer’s team to the vendor.

The Playbook’s Still Being Written

As Chuck mentions in the SVBJ article,

“The playbook hasn’t been written.”

This rings true for all aspects of AI. Simon’s still improving his mission control. Customer Success (CS) teams are still learning what AI governance really means in practice. Vendors offering AI Native Services (AINS) are still figuring out what it means to take responsibility for results at scale.

What’s clear is that we’re moving toward more AI in more workflows, creating more polished results, but making accountability harder to find. The solution—whether it’s a personal mission control, a team standard, or a vendor who owns the outcome—requires a human who has thought carefully about their role.

As the playbook continues to evolve, real success still hinges on having someone willing to own the outcome.

---

🎧 Listen for These Moments

For any leader trying to figure out where personal experimentation ends and institutional AI begins, this one is worth the full listen. Here are three moments to listen for in this week’s episode:

• Strategic AI vs. factual AI: There’s a hard line between using AI for open-ended planning and using it for customer-facing numbers. Simon shares why knowing the difference matters more than most CS leaders realize.

• The comp restructure: Simon shifted his team from 100% GRR-focused comp to a 50/50 GRR/NRR split. What drove that decision, and what it signals about where CS ownership is heading.

• The FDE role and what it signals: Learn where the Field Deployed Engineer sits in Simon’s customer interaction stack, and why he’d always fill that role from within.

---

🔎 Where to Find the Speakers

• Simon Farthing — LinkedIn: https://www.linkedin.com/in/simon-farthing/

• Josh Schachter — LinkedIn: https://www.linkedin.com/in/jschachter/

📎 Referenced in This Episode

• Von — go-to-market intelligence platform

• LoomiAI — Bloomreach’s AI product suite

• Claude — AI assistant and development platform

• Glean — enterprise knowledge management and search

• Gong — revenue intelligence platform

• Granola — AI meeting notes

• “Software companies face existential question as AI changes what customers buy” — Silicon Valley Business Journal, July 2026

---

Wrapping Up

We may be figuring it out as we go, but that doesn’t mean we can’t stop to answer the important questions.

See you next week 🧠