- A growth account executive demonstrates using Claude Cowork to prep for and follow up on customer meetings via custom skills.
- An "account strategy builder" skill pulls scattered data (Salesforce, data warehouse, call recordings, Slack, email, web) into a single account brief in minutes instead of hours.
- After the meeting, a "call transcript processor" skill generates personal action items, an internal Slack summary, and a customer follow-up message — each requiring approval before sending.
Claude Cowork for sales
- Brittany, a growth account executive managing strategic startup accounts, needs to get smart on a customer fast — who they are, what they're building, how they use Claude, their spend, and account risks/growth trends.
- That information is scattered across Salesforce, the data warehouse, call recordings, Slack, email, and the web, and pulling it together used to take hours of manual research.
- To build a skill, she opens a Cowork session and describes what she'd want to know walking into the meeting; Claude drafts a skill file — a plain-text file telling Claude how to approach the task (data sources, usage signals, how to present the analysis).
- In settings she has connected her data warehouse, Salesforce, the web, email, and Slack, and picked a desktop folder where Claude can read, edit, or save files; she tested the skill on a sample account and refined it, and can always open the plain-text file to verify what it does.
- To prep a call, she types "/", picks the account-strategy-builder skill, and names the account; Claude runs all the data pulls at the same time (calls from the last 90 days, revenue trends, open Salesforce opportunities, relevant email/calendar/Slack activity, and funding/web trends).
- The output reads like a full account brief — spend information, stakeholder mapping, which Claude models the customer is building with, open deals, and risk signals — giving her the full landscape before talking to the customer.
- After the meeting, in the same Cowork session (which still has the account context loaded), she picks the call-transcript-processor skill, and Claude generates three things: personal action items, an internal Slack message with key takeaways and next steps/owners, and a customer follow-up message — each shown for approval before sending.
- The post-meeting flow that used to take 30 minutes is done in a couple of minutes and is more thorough than writing from memory; overall she pulls insights from multiple systems and turns them into real account strategy, improving every customer conversation.
Claude Cowork — A Claude workspace where you run sessions and skills that connect to your tools and files.
Skill — A plain-text file that tells Claude how to approach a task (data sources, signals, presentation).
Account strategy builder — A skill that compiles a full account brief from multiple data sources.
Call transcript processor — A skill that turns a meeting transcript into action items, a Slack summary, and a follow-up message.
Account brief — A consolidated document of spend, stakeholders, models used, open deals, and risks for an account.
Connectors — Configured integrations (data warehouse, Salesforce, web, email, Slack) Claude can pull from.
Stakeholder mapping — Identifying the key people involved in an account.
Hi, I'm Brittany. I'm a growth account executive and I manage a portfolio of our most strategic startup accounts.
Before I meet with a customer for the first time, I need to get smart on them fast. Who are they? What are they building? How are they using Claude today? What's their spend look like today on us? And what are the risks and growth trends of this account? Now all of this information exists today, but it's scattered across Salesforce, our data warehouse, call recordings, Slack, email, and the web. Pulling this together used to mean hours of manual research. But with Claude Cowork, I have a skill that does it for me in minutes.
To actually build the skill, I open up a Cowork session and describe what I'd want to know walking into that meeting. Claude drafted a skill file from that conversation, which is basically a text file that tells Claude how to approach the task: What data sources should I be looking at? What usage signals matter? And how to actually present the analysis. In my settings here, I've already connected to our data warehouse, Salesforce, the web, email, and Slack. And I picked a folder on my desktop where Claude can read, edit, or save the files. I ran it on a test account, refined it, and because the file is in plain text, I can always open it and verify what it's doing.
So tomorrow, I've got my first call with a customer. What I do to prep is head into Cowork, type "/", pick the account strategy builder skill, and name the account Acme Corp. Once I enter, Claude will start running all the data pulls at the same time: calls from the last 90 days, revenue trends, open opportunities that exist in Salesforce, any relevant data in my email, calendar, activity, Slack, and any funding trends or things on the web that are most relevant. Then it synthesizes everything into a strategy doc and saves it to my account folder. As you can see, the output will read like a full account brief: spend information, stakeholder mapping, what Claude models the user is building with, open deals, and risk signals. Now I have the full landscape of the account before I've even talked to the customer, and it makes for a way more strategic first conversation. Instead of spending the meeting just getting oriented, I can show up with real context and make the first impression count.
Now, after the meeting, I head back into the same Cowork session, and Claude still has the account context loaded. I type "/", pick the call transcript processor, and Claude will pull the call transcript and generate three things: my personal action items, an internal Slack message with key takeaways, next steps, and action owners, and a customer follow-up message. You can see here that each one comes up for my approval before it actually sends. Now, the whole post-meeting flow that used to take me 30 minutes is done in just a couple of minutes, and it's way more thorough than what I'd write from memory.
With Claude Cowork, I'm pulling insights from multiple different systems and turning them into real account strategy. The quality of every customer conversation is better for it.
TL;DR
- A growth account executive demonstrates using Claude Cowork to prep for and follow up on customer meetings via custom skills.
- An "account strategy builder" skill pulls scattered data (Salesforce, data warehouse, call recordings, Slack, email, web) into a single account brief in minutes instead of hours.
- After the meeting, a "call transcript processor" skill generates personal action items, an internal Slack summary, and a customer follow-up message — each requiring approval before sending.
Takeaways
- Brittany, a growth account executive managing strategic startup accounts, needs to get smart on a customer fast — who they are, what they're building, how they use Claude, their spend, and account risks/growth trends.
- That information is scattered across Salesforce, the data warehouse, call recordings, Slack, email, and the web, and pulling it together used to take hours of manual research.
- To build a skill, she opens a Cowork session and describes what she'd want to know walking into the meeting; Claude drafts a skill file — a plain-text file telling Claude how to approach the task (data sources, usage signals, how to present the analysis).
- In settings she has connected her data warehouse, Salesforce, the web, email, and Slack, and picked a desktop folder where Claude can read, edit, or save files; she tested the skill on a sample account and refined it, and can always open the plain-text file to verify what it does.
- To prep a call, she types "/", picks the account-strategy-builder skill, and names the account; Claude runs all the data pulls at the same time (calls from the last 90 days, revenue trends, open Salesforce opportunities, relevant email/calendar/Slack activity, and funding/web trends).
- The output reads like a full account brief — spend information, stakeholder mapping, which Claude models the customer is building with, open deals, and risk signals — giving her the full landscape before talking to the customer.
- After the meeting, in the same Cowork session (which still has the account context loaded), she picks the call-transcript-processor skill, and Claude generates three things: personal action items, an internal Slack message with key takeaways and next steps/owners, and a customer follow-up message — each shown for approval before sending.
- The post-meeting flow that used to take 30 minutes is done in a couple of minutes and is more thorough than writing from memory; overall she pulls insights from multiple systems and turns them into real account strategy, improving every customer conversation.
Vocabulary
Claude Cowork — A Claude workspace where you run sessions and skills that connect to your tools and files.
Skill — A plain-text file that tells Claude how to approach a task (data sources, signals, presentation).
Account strategy builder — A skill that compiles a full account brief from multiple data sources.
Call transcript processor — A skill that turns a meeting transcript into action items, a Slack summary, and a follow-up message.
Account brief — A consolidated document of spend, stakeholders, models used, open deals, and risks for an account.
Connectors — Configured integrations (data warehouse, Salesforce, web, email, Slack) Claude can pull from.
Stakeholder mapping — Identifying the key people involved in an account.
Transcript
Hi, I'm Brittany. I'm a growth account executive and I manage a portfolio of our most strategic startup accounts.
Before I meet with a customer for the first time, I need to get smart on them fast. Who are they? What are they building? How are they using Claude today? What's their spend look like today on us? And what are the risks and growth trends of this account? Now all of this information exists today, but it's scattered across Salesforce, our data warehouse, call recordings, Slack, email, and the web. Pulling this together used to mean hours of manual research. But with Claude Cowork, I have a skill that does it for me in minutes.
To actually build the skill, I open up a Cowork session and describe what I'd want to know walking into that meeting. Claude drafted a skill file from that conversation, which is basically a text file that tells Claude how to approach the task: What data sources should I be looking at? What usage signals matter? And how to actually present the analysis. In my settings here, I've already connected to our data warehouse, Salesforce, the web, email, and Slack. And I picked a folder on my desktop where Claude can read, edit, or save the files. I ran it on a test account, refined it, and because the file is in plain text, I can always open it and verify what it's doing.
So tomorrow, I've got my first call with a customer. What I do to prep is head into Cowork, type "/", pick the account strategy builder skill, and name the account Acme Corp. Once I enter, Claude will start running all the data pulls at the same time: calls from the last 90 days, revenue trends, open opportunities that exist in Salesforce, any relevant data in my email, calendar, activity, Slack, and any funding trends or things on the web that are most relevant. Then it synthesizes everything into a strategy doc and saves it to my account folder. As you can see, the output will read like a full account brief: spend information, stakeholder mapping, what Claude models the user is building with, open deals, and risk signals. Now I have the full landscape of the account before I've even talked to the customer, and it makes for a way more strategic first conversation. Instead of spending the meeting just getting oriented, I can show up with real context and make the first impression count.
Now, after the meeting, I head back into the same Cowork session, and Claude still has the account context loaded. I type "/", pick the call transcript processor, and Claude will pull the call transcript and generate three things: my personal action items, an internal Slack message with key takeaways, next steps, and action owners, and a customer follow-up message. You can see here that each one comes up for my approval before it actually sends. Now, the whole post-meeting flow that used to take me 30 minutes is done in just a couple of minutes, and it's way more thorough than what I'd write from memory.
With Claude Cowork, I'm pulling insights from multiple different systems and turning them into real account strategy. The quality of every customer conversation is better for it.