⚙️ The future of RevOps

Sales leaders lose visibility the moment reps pick up the phone.

Aircall fixes that.

It gives you real-time insight into every call, every conversation, and every coaching opportunity - without sitting next to your reps.

The result: Better coaching, faster ramp times, and full pipeline visibility.

Teams at Uber, Birchbox, and Sendinblue use it to scale calling without losing control.

👉 Learn more at Aircall.com

How Snowflake rebuilt sales productivity with AI

Denise Persson and Chris Degnan detail how Snowflake transformed marketing and sales using AI, making it operational, secure, and ROI-driven at enterprise scale.

How Snowflake scales AI without chaos

It’s not enough to say “go try AI.” That creates mess, not momentum.

✔ Snowflake formed an AI Council with 30 curious volunteers (across teams) dedicating 20% of their time to test, learn, and roll out new use cases

✔ Every quarter, they host an AI Day to educate the broader org and introduce tools that passed internal governance and security review

➤ CEO buy-in is essential. AI has to be a company priority, not just a side project

This top-down support + structured bottom-up experimentation = scaled adoption with guardrails.

AI as task automation = huge efficiency gains

Chris Degnan kept it simple: AI is a task automator, and that’s what makes it powerful.

✔ Global support saves 418 hours/week
✔ Marketing cut time spent on scripting, prep, and localization by 90%
✔ AI removes busywork so teams can focus

If a tool isn’t saving money or making money, it doesn’t last.

Game-changing models built internally

Snowflake developed two proprietary agents using LLMs and Snowflake Cortex:

➤ Campaign Agent: Delivers real-time ROI for every campaign

⇢ Enables in-quarter resource shifts (e.g. to fix territory pipeline gaps)
⇢ Saves budget, boosts returns

➤ Competitive Intelligence Agent: Custom talking points on demand

⇢ “I’m competing against X for Y use case at Z company” → tailored sales ammo instantly

Without this, enablement at scale would be nearly impossible.

Sales AI is only as good as your data

Snowflake’s mantra: “No AI strategy without a data strategy.”

✔ Snowflake is always customer zero
✔ All internal tools (like Raven) run on governed, centralized, Snowflake-hosted data
✔ Every tool must pass security reviews, even internal experiments

This builds trust, both internally and with enterprise customers wary of LLM misuse.

Sales intelligence without the dashboard chase

Raven is their go-to-market AI assistant. Every rep uses it before customer meetings.

Ask: “What’s the 360 view on Acme Corp?”
Get: Usage trends, risk signals, open tickets, opportunities, blockers - in seconds.

➤ Saves time, yes
➤ But more importantly: it raises the bar on preparation and customer understanding

Even the CEO uses Raven before customer meetings.

Certify your SEs

Chris was asked: “How do you know if your sales engineers are any good?” He didn’t have a clear answer.

So they certified every SE, even the 4th-line leaders.

Then they rolled out Cursor AI: custom demo creation in weeks, not months.

➤ Because the team was technically credible, the AI rollout worked
✘ Without that foundation, automation would have flopped

Org structure matters

Snowflake eliminated data silos by centralizing all analysts under the Chief Data Officer.

Those analysts are embedded in GTM functions but aligned centrally.

➤ This enabled consistent data access, no duplication, and faster AI development

The result: sales, marketing, and execs all look at the same metrics.

Hiring for AI readiness = hire for curiosity

Denise and Chris now prioritize adaptability over resume keywords.

✔ Ask: “How do you learn?” “How do you innovate yourself?”
✔ They hire people who ask questions, explore tools, and adapt fast

➤ This mindset matters more than past experience with tools

And when you’re seen as AI-forward, top talent comes to you.

Mistakes to avoid

1. “Everyone experiment with AI” → chaos
   ⇢ Uncoordinated efforts waste time and duplicate work

2. No ROI? No AI.
   ⇢ Adoption won’t scale if tools don’t save or generate money

3. Siloed teams = stalled AI
   ⇢ Consolidation is critical to move fast with consistency

4. Skipping security and governance
   ⇢ Especially in enterprise, non-compliant tools won’t fly

5. Assuming SEs are technical enough
   ⇢ Certify first. Then deploy tools that assume real skill.

What drives AI success in sales orgs

You need both:

✔ Executive mandate (top-down)
✔ Structured innovation (bottom-up)

If either is missing, progress stalls.

👉 Power up your sales team

If you lead a sales team, run training sessions, or coach reps - this is for you.

With SalesDaily Elite, you get full white-label rights to my complete library of sales cheat sheets, guides, and templates.

✔️ Editable Canva files you can brand as your own

✔️ Ready-to-use for team onboarding, and workshops

✔️ Built for high-impact sales team enablement

Stop wasting time building resources from scratch.

➔ Upgrade to SalesDaily Elite

The real AI unlock is operational rigor

Tessa Whittaker outlines how she’s transforming RevOps at ZoomInfo with AI, by first mastering operational rigor.

Build RevOps like a product org

Before layering in AI, Tessa rebuilt the entire operating model of her team using product management principles:

➤ First step: Understand what you’re working on

She started by asking her team: What are we doing? How do we prioritize? How do we intake requests?

None of it was documented. No roadmap. No clear intake. No capacity model.
So she rebuilt it from scratch, using a simple Google Sheet.

✔ Mapped current projects and guessed at priorities
✔ Estimated team capacity with sprint data
✔ Documented a sprint cadence and standardized releases

➤ Once the manual process worked, she systematized it

She moved the system into Jira, launched a centralized intake form, and created a repeatable process for stakeholder reviews and prioritization.

Key idea: Don’t chase quick wins for the business if your team is operating in chaos. Fix the system first so you can deliver consistently.

Use a maturity scale to measure readiness for AI

Tessa created a 0–5 RevOps maturity scale to benchmark the team’s operational health and decide when AI was appropriate.

Here’s how it works:

0 – Ad hoc and manual
1 – Manual but defined
2 – Systematized rigor
3 – AI-augmented workflows
4 – Agentic assistance (AI supports execution)
5 – AI first, autonomous operations

AI doesn’t enter until level 3.

⇢ You need documented processes, a working system of execution, and strong operational hygiene before AI can add value.

She also built this scale across three other areas:

• Data maturity
• GTM execution maturity
• Operational excellence

You can’t automate what isn’t defined. You can’t add AI to a broken process. And you definitely can’t rely on AI with bad data.

How ZoomInfo is using AI internally (not for sales, for RevOps)

Once her team reached systematized rigor (level 2), they looked at their most time-consuming tasks.

The biggest time sink?

↳ Gathering requirements

Most intake meetings were inefficient. So they built a requirements-gathering agent.

✔ It chats with stakeholders to extract business needs
✔ It asks probing follow-up questions
✔ It rewrites responses into a usable user story
✔ It identifies missing information
✔ It slots directly into Jira

But that’s just the start.

They’re now building a system where the agent predicts sprint load, prioritizes based on goals, and flags delays or over-capacity.

A system for prioritization that scales

To avoid the “loudest voice wins” problem, Tessa implemented centralized stack-ranking:

• Company goals sit at the top
• CRO-level priorities follow
• Then SVP requests get ranked in one central view
• Every stakeholder gets monthly reviews to align

That structure ensures RevOps works above and beyond capacity, but only on what matters most.

And they’re building an AI agent that can auto-score each request based on:

→ Strategic relevance
→ Existing roadmap alignment
→ Run-the-business vs. change-the-business work

She didn’t just teach AI, she made it mandatory

Initially, her team didn’t engage much with AI tools.

So she made one thing mandatory:

✔ Every team member had to build one AI agent
✔ It had to support internal workflows and team productivity
✔ They ran an org-wide hackathon

That’s when things clicked. Once people saw how AI could support their day-to-day tasks, adoption skyrocketed.

This only works because the team is built to scale

Tessa called this out: If you're at a small startup, your RevOps people are generalists. Everyone’s doing intake, documentation, user stories, and stakeholder comms.

At ZoomInfo, each function is specialized:

• Technical systems teams
• Process and product managers
• Business partners for each go-to-market area
• A central operations leader to stitch it together

That separation is what allows scalable execution.

Partnering with:

• Navan: The leading travel and expense platform

• Smartlead: Automate outreach without losing the human touch

• Close: Manage relationships and scale outreach in one smart CRM

• La Growth Machine: Multichannel prospecting that actually scales

(include affiliate links)

Why AI fails in ops (and how to fix it)

Matt Volm and Dominic Freschi break down why most AI initiatives flop, and how to fix them with structure, not just shiny tools.

Why AI keeps failing in RevOps

AI in sales ops sounds exciting. But in reality, most projects fall apart because they’re missing the basics:

✘ No clear goals or KPIs
✘ Dirty, unstructured data
✘ Overhyped expectations from generic AI tools

The result? Random projects, shallow productivity gains, and zero measurable ROI.

When AI makes things harder, not easier

Everyone expects AI to streamline work. But without the right system, it adds friction:

→ Manual prompting with half-baked data
→ Fixing bad or hallucinated outputs
→ Copy-pasting between disconnected tools

Example: Teams use ChatGPT to draft emails, then spend 30 minutes fixing errors and formatting.

Automation ≠ AI

Important distinction:

✔ Automation = Rule-based, predictable workflows
✔ AI = Creative but unreliable suggestions
✔ AI agents = Tools that take action in your systems (e.g. update records, build reports)

If you want scalable, trustworthy AI, you need structure around it.

The 6-pillar framework for AI that works

To move beyond experiments, build these layers around your models:

1. Context orchestration

AI is only as good as the inputs.

• Clean and de-dupe your data

• Filter before you prompt

• Avoid token waste from junk records

Bad data = bad AI. Always.

2. Prompt management

Prompts aren’t one-offs. They’re assets.

• Track versions and test them

• Separate brainstorming prompts from production-ready ones

• Watch for prompt injection risks

Treat prompts like code. That’s how you scale.

3. Model orchestration

No single model wins every task.

• Use different models (GPT, Claude, Gemini) for different jobs

• Rotate to reduce bias and catch errors

• Match strengths to task types

Example: Use Claude for summaries, GPT for ideation.

4. Output validation

Never fully trust an AI output.

• Auto-check for hallucinations or bad links

• Flag strange results for human review

• Compare outputs across models when accuracy matters

AI should support, not sabotage, your workflows.

5. System integration

Insights are worthless if they live in isolation.

• Convert AI outputs into CRM-ready fields

• Push data into Salesforce, Outreach, etc.

• Pull data from your systems to feed the AI

Connect the pipes. Or you’ll be stuck copying and pasting.

6. Track cost and performance

If you’re not measuring, you’re guessing.

• Monitor token and model costs

• Track how often outputs are actually usable

• Measure time saved and actions taken

Otherwise, you’ll waste budget on a black box.

Avoid these traps

✘ Thinking AI will magically reduce workload
✘ Prompting on top of dirty data
✘ Using the same model for everything
✘ Skipping validation or system integration

Don’t scale AI without a structure

Start small, but start right. This framework helps you move from experiments to impact - from shiny object to real business value.

TO-GO

Meredith Chandler: The #1 reason sellers leave a job

Matt Green: So, you cap commissions…

Jason Bay: How to prevent high voluntary turnover rates

Marc Roberge: The future of RevOps