# Ed Marsh

Ed Marsh works on go-to-market and sales improvement for B2B industrial companies, writing about sales leadership, marketing, and practical uses of AI in revenue operations.

- Profile: https://b2binfluencer.marketing/creators/ed-marsh/
- LinkedIn: https://www.linkedin.com/in/edwardbmarsh
- LinkedIn followers: 8,491
- YouTube: https://www.youtube.com/channel/UCFA0TrzwVPYe3WrYH_459Og
- Topics: artificial-intelligence, leadership, sales, revenue-operations, sales-leadership
- Based on 9 dated posts, latest 2026-10-07

## Summary

Ed Marsh is most useful for assessing B2B industrial go-to-market and sales leadership systems.

- Best for: A B2B industrial manufacturer with complex long-cycle sales seeking go-to-market and sales leadership improvement.
- Not for: A fast-moving B2B SaaS or consumer brand seeking broad creator reach and paid placement formats.
- Check first: Confirm audience job titles, seniority, industry, company size, and comparable campaign metrics.

### Takeaways

- Trust and phone selling still matter as discovery moves across social, video, audio, communities, AI answers.
- Most AI speeds up stale tactics; the opening is redesigning the work.
- A CRM-connected dashboard is not reliable if underlying data is incomplete, inconsistent or poorly defined.

## Main ideas

- **The buying journey moved; the inbound playbook did not**: Ed Marsh starts from the view that the fundamentals of B2B selling are unchanged: buyers buy from people they trust, and the phone still works with senior executives. What changed is where attention happens before a buyer talks to sales. They argue the conventional inbound marketing playbook is largely ineffective, websites now serve the bottom of the funnel, and early attention has shifted to AI answers, passively consumed audio and video, scrollable social feeds, and communities. The same shift raises the bar on salespeople: average is not enough, and talent has to be strong in tactical sales skills, business acumen, business finance, and work ethic.
- **Most AI speeds up stale tactics; the opening is redesigning the work**: In Ed Marsh’s account, most AI tools and promises for marketing and sales teams do not move the needle because they make existing playbooks and tactics faster. Creative teams use the same technology differently: to redesign workflows and build capabilities that were essentially unavailable a year earlier. They cite a partnership with Day AI to build go-to-market models across board governance, strategy, product, demand generation, brand and field marketing, revenue operations, complex long-cycle sales, sales management, customer success and after-market selling, and to merge counterproductive silos. The limits are clear: this is not an easy button, it does not replace rigour in understanding buyers, process and methodology, and mistakes are expected.
- **Judge a sales leader by the system, not the record**: Ed Marsh argues that a strong track record is not proof of a great sales leader. The test is whether the leader builds a sales machine that holds up when the market turns or competitors get aggressive. That is why they favour validated assessment over résumé review when hiring for sales leadership, in a post tagged to sales leadership, private equity, industrial sales and go-to-market.
- **A connected AI dashboard is not business understanding**: Ed Marsh pushes back on the claim that wiring an AI assistant such as Claude into CRM and marketing automation means a company now understands how to grow. The reasons are guardrail and context questions, incomplete and non-normalized data, deterministic data conflated with probabilistic output, and missing definitions. They give the approach an “A” for effort while saying business decisions need something better. This marks the boundary: the difference is between rebuilding how work is done and bolting AI onto unready data and undefined terms. They point readers to a LinkedIn Live conversation with Nikos Ntirlis of Databox.

## Work to read

- [B2B buying and selling are changing massively](https://www.linkedin.com/posts/edwardbmarsh_b2b-buying-and-selling-are-changing-massively-activity-7512906178192908288-7pp6)
- [A strong track record isn't proof of a great sales leader](https://www.linkedin.com/posts/edwardbmarsh_salesleadership-privateequity-industrialsales-activity-7511407664925466624-vIaf)
- ["Check out this cool dashboard we built!!"](https://www.linkedin.com/posts/edwardbmarsh_check-out-this-cool-dashboard-we-built-activity-7512924173631995904-YUvI)

## Brand relationships

- **Day AI** (disclosed_partnership)
- **Databox** (disclosed_partnership)

## Partnerships

Ed Marsh's posts do not describe sponsorship or paid partnership formats available to brands. They show working collaborations: Marsh describes partnering with the AI go-to-market platform Day AI to build go-to-market models on it for family-owned industrial companies and middle-market sponsors, and promotes a LinkedIn Live conversation with Nikos Ntirlis of Databox. A third post, about assessing sales leadership candidates, mentions no partner or offer. There is no published partnership product to schedule against, so a brand would need to contact Marsh directly and ask about deliverables, pricing, media-kit specifications, exclusivity terms, application routes, and the format and scope of a potential collaboration. Treat the Day AI and Databox items as examples of the kind of collaboration they engage in, not as an indication of what a sponsorship would cost or include.

## Alternatives

- Rosalyn Santa Elena: Rosalyn Santa Elena is the stronger choice when the manufacturer's real problem sits inside revenue operations rather than at the level of go-to-market strategy: a forecast nobody trusts, stage definitions that differ by manager, commit categories that mean different things to different people, close dates that keep slipping, and deal inspection too thin to surface risk early. The posts lay out a concrete repair path — write the forecast categories down in plain terms, align stage definitions and yes/no exit criteria across the organisation at roughly five or six stages, gate stage movement with required fields, run the forecast on the same cadence each week with the same metrics, track accuracy by rep, manager and segment, pressure-test the numbers against historical stage conversion rates, and build an early warning system for deals that stop moving. A second post makes the business-partner argument directly: the same licence request can be handled as an order taker or as a thought partner who brings allocation data, plan-versus-actual comparison, user-profile review and a recommendation with a next step. If the deliverable the manufacturer needs is a forecast and revenue-operations improvement plan the sales leadership team can run week to week, this is the more specific match.
- Kate Redfern: Kate Redfern is the stronger choice when the manufacturer needs capability built inside the sales organisation rather than a strategic diagnosis of it. The posts describe fractional revenue enablement leadership, sales leadership and coaching, BDR/SDR strategy and roadmap work, manager development programmes, methodology rollouts such as MEDDPICC, Challenger and value selling, onboarding and ramp programmes, QBR design, and AI-enabled coaching systems, alongside reported outcomes including conversion, retention and net new revenue improvements. The second post sharpens the buyer-side argument: executive buyers have already read reviews, spoken to peers and consumed content, so sellers who arrive as if the buying journey is starting from zero waste the meeting. If the manufacturer's plan depends on training managers, rebuilding onboarding, standing up a prospecting function or installing a repeatable methodology, this is the closer match.
- Grace KY Man: Grace KY Man is the stronger choice when the improvement plan will actually be executed through AI agents and the manufacturer needs the operating model designed rather than described. The posts set out a growth loop — a real product signal with a threshold, interpretation that routes clear signals to a rules engine and messy ones to a larger model, an action written to the exact account and behaviour rather than a template blast, and a human gate on anything that cannot be undone — plus guardrails agreed before the agent runs: a per-action cost ceiling and model tier with human handoff past the cap, data scoped tier by tier, and an explicit list of which actions the agent may take alone. The second post shows the same thinking applied to a taught programme, with a 90-day execution plan, live demos and governance, and student reports of turning agent work into an operating system rather than a set of tools. If the plan's centre of gravity is a governed agentic workflow across research, content, outbound and CRM, this fits more closely.

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