# Vernon Keenan

Vernon Keenan writes analysis about Salesforce and enterprise AI, including AI governance, Salesforce’s changing interface, and the tools used to manage enterprise systems.

- Profile: https://b2binfluencer.marketing/creators/vernon-keenan/
- LinkedIn: https://www.linkedin.com/in/vernonkeenan
- LinkedIn followers: 34,453
- Topics: artificial-intelligence, ai-agents
- Based on 11 dated posts, latest 2026-09-23

## Summary

Most useful for analysis that connects Salesforce and enterprise AI announcements to who controls agent context, permissions, and infrastructure.

- Best for: A B2B SaaS platform selling enterprise AI governance or agent-control tooling for a thought-leadership series.
- Not for: Short-form video, scripted product demos, or nontechnical HR and payroll campaigns.
- Check first: Confirm rates, deliverable formats, editorial control, exclusivity, disclosure, and recent audience analytics.

### Takeaways

- Enterprise AI is a control race, not an interface race.
- AIforce coverage links events across 168 days and separates adoption from accuracy evidence.
- Harness scorecard separates delivered, partial, and open capabilities, including security and FinOps gaps.

## Main ideas

- **Enterprise AI is a control race, not an interface race**: Keenan argues that when Salesforce moves work into Claude, Slack, Lightning, or wherever people and agents work, the visible screens are not the product. The governed context underneath—who can see what, how agents are constrained, how permissions hold—is the real competitive ground. The interface shift matters only if the Trusted Enterprise AI Harness and control layer actually ship. Without that, the UI move is a costume.
- **Category naming as strategy: Delivery Digital Twins and control without data gravity**: Keenan treats acquisitions as signals about which technical assets become governance infrastructure. They call the capability class a Delivery Digital Twin—a dependency-resolved model of a Salesforce org’s objects, fields, automations, permissions, code, and dependencies that AI agents can query before changing anything. The strategic move is “control without data gravity”: govern agent work across enterprise applications while business records stay where they are.
- **Test vendor claims against quality evidence, not adoption metrics**: Keenan prioritizes evidence that shows whether a claim is true. When Salesforce says Claude works better through AIforce than alone, the proof needed is a with-and-without comparison on correctness and permission boundaries, not usage counts. This is a guest analysis by Chris Pearson, presented through Keenan’s account; credit the argument to Pearson.
- **Score promises against delivery, and name the gaps**: Keenan uses a delivery scorecard to judge enterprise AI announcements: what was promised, what shipped, what is partial, and what remains open. They quote executives acknowledging gaps instead of smoothing them over and compare competitors already shipping similar control planes. A console that meters spend covers only part of the FinOps job, not the whole governance answer.

## Work to read

- [Salesforce AIforce and the 168-day interface story](https://www.linkedin.com/posts/vernonkeenan_salesforce-aiforce-dreamforce-activity-7505683165970055172-IEqL)
- [The $400 million admin tool: ServiceNow, Sweep, and Delivery Digital Twins](https://www.linkedin.com/posts/vernonkeenan_salesforce-servicenow-salesforcedevops-activity-7503110077428531200-X1XT)
- [Salesforce says Claude works better with AIforce — where’s the proof?](https://www.linkedin.com/posts/vernonkeenan_salesforce-aiforce-agentforce-activity-7508188640112427008-CgMI)
- [Salesforce named the enterprise AI harness. Now it has to build it.](https://www.linkedin.com/posts/vernonkeenan_salesforce-agentforce-enterpriseai-activity-7503867157718437888-QEZG)

## Brand relationships

- **SalesforceDevops.net** (mention)
- **Salesforce** (mention)
- **ServiceNow** (mention)

## Partnerships

None of Vernon Keenan’s four posts documents a sponsorship or partnership offer. Each analyzes Salesforce and enterprise AI topics and points readers to a longer article, several on SalesforceDevops.net, including one by a guest author credited by name. This shows they work with outside contributors and publish beyond LinkedIn, but does not establish whether sponsored placements are available. Before budgeting, ask them directly about rates, formats, deliverables, editorial control, exclusivity, disclosure practices, and how to open a sponsorship conversation.

## Alternatives

- Felix Markman: Choose Markman when the campaign should focus on adoption risk and payback rather than tracking vendor announcements. The posts argue that “AI replaces the UI” is a keynote line and that, a year on, well under 20% of users will be independent of logging in; that demos always look finished while adoption remains the real barrier; and that the hardest business problems yield to clean data, solid governance, and systems that earn user trust. For the same governance-themed series, Markman is the better pick for a skeptical, ROI-first counterweight aimed at a revenue-operations audience. They argue that putting an unproven surface on production CRM costs a year of cleanup and a hit to trust, not a refund. That makes them a natural fit for a launch piece or webinar that asks buyers to slow down and measure, while positioning the vendor’s product as the unglamorous foundation that must come first.
- Mike Coutts: Choose Coutts when the campaign needs the buyer-side operational view rather than vendor strategy. The posts describe what a tenth Dreamforce looked like inside companies running commercial operations: data still messy, now with a layer that can be confidently wrong about it; reps still buried in administration; change management the line item most often cut; and the pace of change as a problem in its own right. Coutts’s central argument—that the Manual Tax is where the revenue is, and that placing an agent on top of a broken process automates the tax rather than removing it—gives a governance campaign a concrete reason to fund process and data work before agents. The cost-of-delay post makes the same case commercially, describing a manufacturer running quote-to-cash across an ERP, a CRM the sales team works around, and about 40 spreadsheets. The bill, Coutts argues, tends to arrive in diligence or a value-creation review, when there is no runway left. That suits a practitioner webinar for an audience that must justify the spend internally and needs a small, measurable first use case.
- Shubham Goel: Choose Goel when the campaign needs an enterprise IT buyer’s questions rather than an analysis of vendor material. The posts synthesize issues raised by IT executives at large banks, telecom, and technology companies: who governs the apps teams now build themselves, who owns AI risk, whether oversight belongs in IT or an independent body, employees recast as managers of agent teams, documentation gaps that make agent output unusable, and CRM–ERP integration as the underlying blocker. For the same campaign, Goel is the strongest of the three alternatives if the goal is to test the vendor’s governance message against what a CIO audience actually asks—or to add an outside signal that AI adoption is uneven, drawn from people who see many enterprises rather than the vendor’s own references. The post noting that Salesforce is presenting itself as an AI company and asking whether the balance of power shifts when the screen changes comes closest to the series’ framing of any alternative item, from a buyer-side rather than seller-side position.

---
AI-assisted summary of public posts. Audience demographics are not available. Engagement appears only where we checked recent posts. Method: https://b2binfluencer.marketing/methodology/