We’re thrilled to announce that Surveys Agent - your intelligent assistant for managing surveys end-to-end - is now generally available to all commercial customers worldwide with a Microsoft 365 Copilot license. 

Previously available through the Frontier preview program, Surveys Agent has helped early adopters streamline data collection and actionable insights for use cases like training sessions, employee engagement, and customer research. Now, it’s ready for everyone. 

What can Surveys Agent do? 

Surveys Agent simplifies the entire survey lifecycle through a conversational experience. Whether you're starting from scratch or refining an existing survey, it helps you: 

• Create surveys instantly in a shareable Microsoft Forms draft 

• Refine your survey to improve question clarity and structure 

• Plan launch timelines and distribution strategies 

• Send invitations and reminders to maximize responses 

• Monitor progress and keep you apprised via Outlook 

• Get status, summary of response and open in Excel for deep analysis 

No more juggling multiple tools or chasing down responses. Surveys Agent brings everything together in one place, making survey creation as easy as chatting with an expert colleague. 

Built into Microsoft 365 Copilot 

From drafting to distribution, Surveys Agent works seamlessly inside your Copilot experience. It suggests optimal launch plans, prepares outreach messages, and keeps you updated with response tracking. You can revisit the conversation anytime to check status, ask follow-up questions, or export results to Excel for deeper analysis. 

What’s new with General Availability? 

With this broader release, Surveys Agent is now accessible to all commercial Microsoft 365 Copilot users. You can find it in the Agent Store under Built by Microsoft, ready to help you build, run, and analyze surveys with less effort. 

With general availability, we’ve also introduced several key enhancements to make Surveys Agent even more powerful and intuitive: 

• Grounding to your content: You can now connect Surveys Agent to Word, Excel, and PowerPoint files, as well as existing Microsoft Forms survey links, to help draft and refine surveys with richer context. 

• Interactive tutorial prompt: Learn about Surveys Agent’s capabilities in more depth before diving in—perfect for first-time users. 

Try Surveys Agent today 

Surveys Agent is available to business users with a Microsoft 365 Copilot license. Ready to get started? Install Surveys Agent and pin it to your Microsoft 365 Copilot sidebar for quick access. We’re excited to see how you use it—and we’re listening. Share feedback directly in the chat by rating responses with a thumbs up or thumbs down and adding your comments to help us shape what’s next. 

Looking forward to seeing how Surveys Agent can support your workflow and continue to bring you from idea to insights with less effort! 

three things MCP can do, and an infinite number of things it can’t do (all of which make it great)

AI agents can interact with the world using tools. Those tools can be generic or specific.

The most general ones, like “run a bash command” and “read and write files” are built into the agent. More specific ones are provided through Model Control Protocol (MCP) servers.

Every tool provided to the agent comes with instructions sent as part of the context. Each MCP server the user configures clogs up the context with instructions and tool definitions, whether the agent needs them for this conversation or not.

If the agent can run a bash command, it can write a curl command or a script to call an API. Why use an MCP server instead?

MCPs provide three unique abilities.

1. Authentication. Authorize an MCP server once to act as you, and then take many actions, each properly attributed. OAuth is hard and you can’t do it with curl. (OK, it’s more than once, it’s ‘every time it loses the connection’. This feels like every day, but maybe the agents will get better at renewing auth.)
2. Specialized interface. A software API is optimized to talk to code. If it responds with JSON, that is verbose and uses a ton of tokens. An MCP response can summarize the results in text. It can intersperse that with CSV and even ASCII art! It’s more efficient and effective in communicating with an LLM.
3. Change. MCPs don’t have to be consistent from day to day, since every conversation is new. The creators of an MCP server can work on that response and make it more effective, changing its format at need. They can add tools, change tools, and even remove tools that aren’t used enough. Try doing that in a software API! It’d break every program that uses it. MCPs can iterate, and rapid iteration is a superpower that AI gives us.

If you want to teach your agent to do something that doesn’t require authentication–like read a web site–then by all means, let it use the tools it already knows. It can get a long way with `curl` and `jq`. Why dilute its world with more instructions when it already knows so much?

MCPs don’t let the agent do anything else.

While “run a bash command” covers most things you want it to do, it also covers everything you don’t want an agent to do. The agent can screw with your configuration, write private data out to a public repository, and use your credentials to publish infected versions of your libraries. There is (relative) safety in specific tools. For instance, the agent’s filesystem tools reject writes to files outside of the current project. (The agent then asks my permission to do that update in bash. I say no.)

Well-designed MCPs offer the operations that make sense. They’re limited by your authorization as a user, and you can further limit their authorization when you connect or in your agent’s configuration. We can be smarter about it.

Someday we will have nice things.

Currently, if I configure an MCP, it’s available all the time to all agent threads. Most of the time, that’s a waste of my context. I want to configure which subagents know about which MCP, so my “look at production” agent can see my observability platform, my UI-updating agent can see Figma, and my status update agent can see Asana. I also want agents to load MCP context incrementally, so that it doesn’t get every tool definition until it asks to see them.

When MCPs don’t hog context, they still won’t often beat using the innate knowledge of the model. But when you are ready to curate the access that agents have to your SaaS or data, MCPs are fantastic.

Cory House (Pluralsight/DomeTrain author and principal at ReactJS Consulting) shares the story of going “all-in” on JavaScript/React and how that focus grew into a successful independent consulting and training career. We dig into the tradeoffs of deep specialization vs breadth, how to spot real opportunities, and the “two-way door” idea for tech career moves. Cory also walks through his current pivot: using AI as a developer accelerator (how teams use it, where it helps most, what to watch out for) and how experimentation today — while tooling is cheap and rapidly evolving — is valuable. Along the way we surface mindset lessons (Cal Newport, Carol Dweck), how to balance giving away content vs paid courses, and practical tips for auditors/consultants trying to scale their impact.

Guest: Cory House — https://www.bitnative.com/

· Consulting & training: https://www.reactjsconsulting.com/

· Courses: Dometrain (TypeScript: Getting Started / Deep Dive) · YouTube: https://www.youtube.com/@housecor

· X: https://x.com/housecor

· GitHub: https://github.com/coryhouse

· DevOpsDays Des Moines (speaker): https://devopsdays.org/events/2025-des-moines/welcome/

· Podcast: https://eitl.ai/podcast/

· Books: So Good They Can’t Ignore You (Cal Newport), Mindset (Carol Dweck)

Download audio: https://anchor.fm/s/104ead10c/podcast/play/110922361/https%3A%2F%2Fd3ctxlq1ktw2nl.cloudfront.net%2Fstaging%2F2025-10-9%2F412136523-44100-2-5126169d8ed8f.mp3

As AI agents become more capable, there's growing eagerness to develop long-running tasks that operate autonomously with minimal human intervention. However, my experience suggests this fully autonomous mode doesn't apply to most documentation work. Most of my doc tasks, when I engage with AI, require constant iterative decision-making, course corrections, and collaborative problem-solving—more like a winding conversation with a thought partner than a straight-line prompt-to-result process. This human-in-the-loop requirement is why AI augments rather than replaces technical writers.

---

Jen Abel is GM of Enterprise at State Affairs and co-founded Jellyfish, a consultancy that helps founders learn zero-to-one enterprise sales. She’s one of the smartest people I’ve ever met on learning enterprise sales, and in this follow-up to our first chat two years ago (covering the zero to $1 million ARR founder-led sales phase), we focus on the skills founders need to learn to go from $1M to $10M ARR.

We discuss:

1. Why the “mid-market” doesn’t exist

2. Why tier-one logos like Stripe and Tesla counterintuitively make the best early customers

3. The dangers of pricing your product at $10K-$20K

4. Why you need to vision-cast instead of problem-solve to win enterprise deals

5. Why services are the fastest way to get your foot in the door with enterprises

6. How to find and work with design partners

7. When to hire your first salesperson and what profile to look for

—

Brought to you by:

WorkOS—Modern identity platform for B2B SaaS, free up to 1 million MAUs

Lovable—Build apps by simply chatting with AI

Coda—The all-in-one collaborative workspace

—

Where to find Jen Abel:

• X: https://x.com/jjen_abel

• LinkedIn: https://www.linkedin.com/in/earlystagesales

• Website: https://www.jjellyfish.com

—

Where to find Lenny:

• Newsletter: https://www.lennysnewsletter.com

• X: https://twitter.com/lennysan

• LinkedIn: https://www.linkedin.com/in/lennyrachitsky/

—

In this episode, we cover:

(00:00) Welcome back, Jen!

(04:38) The myth of the mid-market

(08:08) Targeting tier-one logos

(10:50) Vision-casting vs. problem-selling

(15:35) The importance of high ACVs

(20:45)  Don’t play the small business game with an enterprise company

(25:09) Design partners: the double-edged sword

(28:11) Finding the right company

(36:55) Enterprise sales: the art of the deal

(43:21) The problem with channel partnerships

(44:41) Quick summary

(50:24) Hiring the right enterprise salespeople

(56:49) Structuring sales compensation

(01:01:01) Building relationships in enterprise sales

(01:02:07) The art of cold outreach

(01:07:31) Outbound tooling and AI

(01:14:08) Lightning round and final thoughts

—

Referenced:

• The ultimate guide to founder-led sales | Jen Abel (co-founder of JJELLYFISH): https://www.lennysnewsletter.com/p/master-founder-led-sales-jen-abel

• Mario meme: https://www.linkedin.com/pulse/missing-meme-led-me-woman-johann-van-tonder-im6df

• Kathy Sierra: https://en.wikipedia.org/wiki/Kathy_Sierra

• Cursor: https://cursor.com

• The rise of Cursor: The $300M ARR AI tool that engineers can’t stop using | Michael Truell (co-founder and CEO): https://www.lennysnewsletter.com/p/the-rise-of-cursor-michael-truell

• Justin Lawson on X: https://x.com/jjustin_lawson

• Stripe: https://stripe.com

• Building product at Stripe: craft, metrics, and customer obsession | Jeff Weinstein (Product lead): https://www.lennysnewsletter.com/p/building-product-at-stripe-jeff-weinstein

• He saved OpenAI, invented the “Like” button, and built Google Maps: Bret Taylor on the future of careers, coding, agents, and more: https://www.lennysnewsletter.com/p/he-saved-openai-bret-taylor

• OpenAI’s CPO on how AI changes must-have skills, moats, coding, startup playbooks, more | Kevin Weil (CPO at OpenAI, ex-Instagram, Twitter): https://www.lennysnewsletter.com/p/kevin-weil-open-ai

• Anthropic’s CPO on what comes next | Mike Krieger (co-founder of Instagram): https://www.lennysnewsletter.com/p/anthropics-cpo-heres-what-comes-next

• Linear: https://linear.app

• Linear’s secret to building beloved B2B products | Nan Yu (Head of Product): https://www.lennysnewsletter.com/p/linears-secret-to-building-beloved-b2b-products-nan-yu

• Gemini: https://gemini.google.com

• Microsoft Copilot: https://copilot.microsoft.com

• How Palantir built the ultimate founder factory | Nabeel S. Qureshi (founder, writer, ex-Palantir): https://www.lennysnewsletter.com/p/inside-palantir-nabeel-qureshi

• McKinsey & Company: https://www.mckinsey.com

• Deloitte: https://www.deloitte.com

• Accenture: https://www.accenture.com

• Building a world-class sales org | Jason Lemkin (SaaStr): https://www.lennysnewsletter.com/p/building-a-world-class-sales-org

• Peter Dedene on X: https://x.com/peterdedene

• Hang Huang on X: https://x.com/HH_HangHuang

• Hugo Alves on X: https://x.com/Ugo_alves

• A step-by-step guide to crafting a sales pitch that wins | April Dunford (author of Obviously Awesome and Sales Pitch): https://www.lennysnewsletter.com/p/a-step-by-step-guide-to-crafting

• Clay: https://www.clay.com

• Apollo: https://www.apollo.io

• Jason Lemkin on X: https://x.com/jasonlk

• Gavin Baker on X: https://x.com/GavinSBaker

• Jason Cohen on X: https://x.com/asmartbear

• Baywatch on Prime Video: https://www.primevideo.com/detail/Baywatch/0NU9YS8WWRNQO1NZD5DOQ3I8W6

• Playground: https://www.tryplayground.com

• ClassDojo: https://www.classdojo.com

• Jason Lemkin’s post about Replit: https://x.com/jasonlk/status/1946069562723897802

—

Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email podcast@lennyrachitsky.com.

—

Lenny may be an investor in the companies discussed.

To hear more, visit www.lennysnewsletter.com



Download audio: https://api.substack.com/feed/podcast/177909982/7930cad72c68cf7f2bb4cb5598843995.mp3