ship these 5 claude code workflows this week

Over the past few months, we’ve been using Claude Code across our GTM operation at RevyOps with a team of 9 developers testing different workflows and implementations.

Some workflows were useful in isolated cases. A smaller number proved reliable enough to become part of our day-to-day operation.

These are the 5 workflows we decided to keep.

Each one builds on the previous workflow, so it’s worth starting from Workflow 1 rather than jumping ahead.

The long-term value comes from the foundation and how the workflows connect together over time.

Workflow 1: CRM cleanup

This takes 10 minutes and it should be the first thing you do before touching anything else.

cd ~/claude-gtm
claude

Read contacts.csv and icp-context.md.
Deduplicate by email address.
Standardize all job titles into consistent categories.
Flag rows where email domain is gmail, yahoo, hotmail,
or any personal provider.
Remove any contact that does not match ICP criteria
in icp-context.md.
Save as contacts-cleaned.csv with a removal log
showing what was removed and why.

Why this matters: everything you build after this runs on this data. If the data is dirty, every downstream workflow underperforms. Clean it once. Run this same prompt on every new export. The .md file keeps the criteria consistent.

Workflow 2: Transcript analysis for ICP scoring

If you have call recordings from Gong or Fathom, drop the transcripts into your project folder as text files. If you do not have transcripts yet, skip to Workflow 3 and come back to this when you do.

claude

Read every file in /transcripts.
Analyze across all calls and identify:
- Top 5 objection patterns with exact language buyers use
- Win signals: what was present in deals that closed
- Loss signals: what was present in deals that stalled
- Title, industry, and company size patterns across wins
- Pain point language buyers use to describe their problems

Output an ICP scoring framework based on these patterns.
Format as a markdown file I can save as
icp-scoring-criteria.md.

Take that output and update your icp-context.md with the patterns it found. Now your ICP criteria is built from revenue data instead of assumptions. Every workflow that reads icp-context.md after this point scores against real patterns.

Workflow 3: Bulk enrichment via direct API

This is where you start spending tokens so the routing matters. Direct API for volume. MCP for filtering.

claude --plan

Read contacts-cleaned.csv, icp-context.md,
and data-sources.md.

Step 1: Enrich each contact using Apollo direct API.
Pull firmographic data, tech stack, funding status,
headcount, and hiring patterns.
Do NOT use MCP for this step. Direct API only
for token efficiency on bulk processing.

Step 2: Score each contact against the ICP scoring
criteria in icp-context.md. Bucket into:
- Tier 1: high fit + strong signal
- Tier 2: good fit, weaker signal
- Tier 3: low fit or no signal

Step 3: Filter via Gmail MCP — remove any contact
I have already emailed.

Step 4: Filter via Stripe MCP — remove any contact
who is an existing customer.

Save as scored-leads.csv with columns for score,
tier, enrichment data, and scoring rationale.

Use Plan Mode. Show me the architecture before
building. Pause before final save for my review.

Review the scored-leads.csv. If the tiering does not match your intuition about which leads are actually good, refine your icp-context.md and run again. Second run is always sharper.

Workflow 4: Sequence generation from scored leads

Make sure your messaging-logic.md file is in the project folder. If you have not built it yet, go back to the Day 4 guide and build it first. This workflow depends on it.

claude

Read scored-leads.csv, icp-context.md,
and messaging-logic.md.

For each Tier 1 lead, generate a 3-email sequence:
- Each email under 70 words
- Soft question-based CTA in email 1
- Lead with the messaging angle mapped to their
  ICP segment in messaging-logic.md
- Include a front-end offer (audit, playbook,
  calculator, or teardown) based on their pain points
- Generate 3 variations per email:
  simple, niche-aware, hyper-specific
- Clay-compatible merge variables:
  {{first_name}}, {{company}}, {{pain_point}},
  {{tech_stack}}
- Spintax subject lines (3 variations each)

For Tier 2 leads, generate a 2-email templated
sequence with less personalization.

Drop Tier 3 entirely.

Output as sequences-ready.csv formatted for
Instantly import.

Pause before final export for my review.

Review at the checkpoint. If the messaging sounds off for a specific segment, update messaging-logic.md and run again. The logic improves. The output improves. That is the loop.

Workflow 5: Pipeline reporting via Cowork

This is where Cowork earns its place. Code built the pipeline. Chat helped design the strategy. Now Cowork packages the results into something your team or clients can actually use.

Open Claude Cowork. Point it to your project folder. Run this:

Read scored-leads.csv and the most recent
sequences-ready.csv in this folder.

Build an Excel pipeline report with the
following structure:
- Tab 1: Tier 1 leads with scores, enrichment
  data, and assigned sequence variant
- Tab 2: Tier 2 leads with scores and
  templated sequence assignment
- Tab 3: Summary dashboard — total leads by tier,
  average ICP score, top industries represented,
  top titles represented
- Tab 4: Campaign readiness checklist — leads
  ready to send, leads needing review, leads dropped

Use conditional formatting on ICP scores.
Color code tiers.
Save to /output as pipeline-report.xlsx.

Schedule this to run weekly on Monday mornings. Every week your team gets a fresh pipeline report without anyone building it manually. That is Cowork operating as the formatting layer on top of Code's infrastructure.

Do one workflow per day starting today. By next Wednesday you have a connected pipeline running from clean CRM data to scored leads to personalized sequences to a formatted Excel report.

The revenue is in the engineering.

Stay safe out there.

Much Love,

Benjamin Reed

Revyops & NextGen Founder

P.S. Check out ReyvOps at Revyops.com.

P.P.S. Want to help me on my journey to build RevyOps into a $1 Million SaaS? Please share this newsletter with your network so they can follow along. Anyone can sign up 100% free using this link: Click here

The Builder's Playbook is a series by Benjamin Aaron Reed on the tools, systems, and workflows that give founders and GTM operators an unfair advantage. New issues drop regularly.