What AI tools do you use?

We build on top of the tools you already have - HubSpot, Salesforce, n8n, Make - and layer in AI from OpenAI, Anthropic, and others where it makes sense. No proprietary platform, no new vendor to manage.

Is this actual AI or just rule-based automation?

Both, depending on the job. Simple routing and alerts are rule-based - that's the right tool for that. Scoring, summarization, intent detection, and research use real LLMs. We match the approach to the problem.

Do I need to replace my CRM?

No. We build on top of what you have. Whether that's HubSpot, Salesforce, or something else, we connect to it - we don't replace it.

Do I need a technical team to manage this?

No. The automations run on their own. We build it, hand it off, and your team uses the outputs - alerts, scores, summaries - without touching the infrastructure.

How do you handle our data?

Your data stays in your stack. We don't store it, sell it, or route it through proprietary systems. The pipelines we build connect your existing tools directly.

What does onboarding actually look like?

A 30-minute discovery call, then we scope the work and send a proposal. Once you're in, we go heads-down - no lengthy kickoffs or status meetings. You see working software fast.

What if the AI gets it wrong?

It will sometimes, especially early. That's why we build human review steps where it matters and tune based on your data. Accuracy improves over time - it's not set-and-forget.

How is this different from HubSpot's or Salesforce's native AI?

Native AI works within the walls of that one tool. We connect your entire stack and build logic that runs across it - enrichment, scoring, alerts, and summaries that pull from multiple sources and trigger actions anywhere.

How to Set Up Claude Cowork for Sales Ops (Step by Step)

Claude Cowork is not a chat interface. It’s a desktop agent - you point it at a folder, give it context about your business, and it executes tasks autonomously. For sales ops teams, that means scheduled pipeline reviews, automated CRM updates, and call prep briefs that write themselves. Here’s how to set it up.

What You Need Before You Start

Cowork requires a paid Claude plan (Pro at $20/month or Team at $30/user/month) and the Claude Desktop app on macOS or Windows. It’s not available on the web or mobile.

Once installed, the desktop app has three tabs at the top: Chat, Cowork, and Code. Click Cowork.

That’s the whole setup. The rest is configuration.

Step 1: Create a Working Folder and Load It With Context

Cowork works by reading files in a folder you designate. Every session in that folder starts with Claude reading those files - so anything you put there becomes persistent context.

Click “Work in a Folder,” select your folder, and grant access. Then create three files inside it:

about-us.md - Your company, your CRM setup, your sales process, your team structure. Be specific. “We use HubSpot. Deals move through five stages. Rep team is seven AEs. Quota is $80K per quarter.”
working-rules.md - How you want output formatted. File naming conventions. What to skip. “Always save output as a .md file. Never include a summary section. Flag deals over $50K separately.”
pipeline-context.md - A running file you update weekly with current pipeline state, deal notes, and priorities. Claude reads it at the start of every session.

This folder is your ops brain. The better the context files, the less you re-explain every session.

Step 2: Set Global Instructions

Go to Settings > Cowork > Global Instructions. Write standing instructions that apply to every Cowork session, regardless of folder.

For a sales ops setup, include: your role and what you’re responsible for, which tools you use (HubSpot, Salesforce, Gong, Slack), and any consistent formatting rules. This is also where you set tone - “Write like a practitioner, not a consultant. Short sentences. No filler.”

Global instructions load before your folder context. Think of it as the brief you’d give a new ops hire on day one.

Step 3: Set Up a Scheduled Pipeline Brief

This is where Cowork earns its keep. Type /schedule in any task prompt, or click “Scheduled” in the left sidebar and create a new task.

Set it to run weekdays at 7am. The task: pull your pipeline file, identify deals with no activity in the last 14 days, flag anything closing this month with a probability below 50%, and write a brief with the three deals that need attention today. Save it as pipeline-brief-[date].md.

That file is waiting in your folder when you open your laptop. No prompt, no copy-paste, no dashboard to check.

The catch: your computer needs to be on and the app needs to be open when the task fires. If the machine is asleep, Cowork queues it and runs as soon as the app opens.

Step 4: Connect Your CRM

Cowork has native connectors for Salesforce, HubSpot, Apollo, Clay, and Outreach. Connect the ones your team uses through Settings > Plugins.

Once connected, Cowork can read and write to your CRM directly - no browser required. You can ask it to pull all deals updated in the last 48 hours, identify the ones missing MEDDIC fields, and draft update notes based on the last Gong call transcript. It writes those notes back to the deal record.

If your CRM isn’t on the connector list, Cowork falls back to browser - it opens Chrome, navigates to your CRM, and updates records by clicking and typing through the UI. Slower, but it works.

Step 5: Build a Reusable Skill for Recurring Workflows

For anything you do more than once a week, encode it as a Skill. A Skill is a saved set of instructions Cowork loads on command - so you type /prep and it runs your full pre-call brief workflow without re-explaining the steps.

Build one for call prep: pull the contact from HubSpot, find their last three Gong call summaries, check their deal stage and open tasks, and write a one-page brief. Save it. Every rep on the team can run it with one command.

Skills are what separate Cowork from a chat interface. Chat requires re-prompting. Skills run.

One Limitation Worth Knowing

Cowork does not have automatic persistent memory between sessions. Context resets. The workaround: maintain a summary.md file in your folder and tell Claude to update it at the end of each session. It acts as compressed memory - deal status, decisions made, open questions - that loads fresh the next morning.

It’s a manual system for now. Worth building.

The teams that get the most out of Cowork are the ones that treat it like an ops hire - give it good context upfront, build repeatable workflows, and let it run. The ones who don’t are the ones who keep prompting it like a chatbot.