01
CRM cleanup
Fields, stages, ownership, duplicates, missing values, inconsistent records, and import preparation.
CRM & workflow implementation
I fix the messy layer between your CRM, workflows, and connected tools.
I help HubSpot, RevOps, CRM, and agency teams with routing, field mapping, workflow fixes, integrations, reporting cleanup, QA, and handoff notes.
Small fixed scopes
One clear workflow, integration, routing problem, or cleanup block.
Review-first when needed
Ambiguous records go to humans instead of being guessed into the CRM.
Implementation, not decks
The output is a working system and a usable handoff.
What I actually help with
The practical work behind a CRM that people can trust.
I am usually brought in after the business direction is clear and the implementation still needs to be built, fixed, tested, or cleaned up.
02
Workflow repair
Broken triggers, bad filters, missing actions, partial failures, retries, and workflows nobody trusts.
03
Routing & field mapping
Lead owners, booked-call assignment, CRM object mapping, safe matching keys, and fallback rules.
04
Integration glue
Make, Zapier, n8n, APIs, webhooks, payload cleanup, record creation, and connected-tool sync.
05
Reporting cleanup
Fixing the CRM inputs and workflow states that make reports incomplete, inconsistent, or misleading.
06
QA & handoff
Real-record testing, edge cases, dedupe checks, review queues, Loom videos, and implementation notes.
Selected implementation work
Anonymized examples from real workflow problems.
Most client systems are private. These examples show the actual problem, what was built, and the operational result without exposing client data.
Pipedrive enrichment
Review-first AI enrichment for CRM leads
Built for a franchise advisory team that needed cleaner lead research without allowing AI to overwrite live CRM data blindly.
Pipedrive
Make
OpenAI
Google Sheets
Problem
New leads still required repeated manual research before the records were useful.
Built
A label-triggered enrichment flow with sourced research, structured output, and manual review.
Outcome
The workflow went live, was accepted and paid, and led to another requested integration.
Recruiting & client operations
Fireflies, Salesforce, Lever, Slack, and AI notes
A group of n8n workflows for a staffing company with meeting, candidate, CRM, ATS, and account-channel activity spread across several tools.
n8n
Salesforce
Lever
Fireflies
Slack
Problem
Important calls, candidate interviews, transcripts, and CRM emails were not reaching the right place.
Built
Meeting matching, CRM logging, ATS notes, transcript files, Slack routing, ledgers, and review paths.
Outcome
Production workflows created usable records while ambiguous matches were routed for human review.
Pipedrive booking workflow
Booked calls routed into the correct pipeline and owner
Built for an outbound agency where bookings, ownership, pre-call emails, and subscriber sync were fragile or manual.
Pipedrive
Cal.com
Make
Beehiiv
Problem
Booked meetings did not consistently create the correct CRM records or assign the right rep.
Built
Person and deal creation, pipeline placement, rep assignment, pre-call emails, and subscriber sync.
Outcome
The booking flow, routing logic, email sequence, and Beehiiv sync were tested and working.
Internal systems proof
Gmail-to-Airtable revenue control layer
Built inside an internal CAD/BIM software sales operation where lead, quote, order, payment, and follow-up status lived across inboxes and spreadsheets.
Gmail
Airtable
Make
OpenAI
Brevo
Problem
The inbox showed activity, but there was no reliable structured state for next actions and ownership.
Built
Lead, contact, task, quote, event, review, payment, and QA workflows around Gmail and Airtable.
Outcome
The system moved into daily internal use and made revenue-related workflow state visible outside Gmail.
How I deliver
Small enough to ship. Safe enough to trust.
The goal is not to make the automation look clever. The goal is to make the process predictable, testable, and understandable after handoff.
Find the real failure
Review the current workflow, records, payloads, fields, and examples instead of relying only on the intended process.
Define safe rules
Confirm the source of truth, matching keys, routing logic, overwrite rules, and cases that need review.
Build and test
Ship the smallest reliable version and test happy paths, duplicates, missing data, bad access, and partial failures.
Hand it over
Document the trigger, fields, logic, edge cases, known limitations, and how the team should operate it.
Best fit
Where I fit into the team.
I work best as the practical implementation layer—not as a replacement for the senior strategist who owns the overall RevOps or GTM direction.
Good fit
- HubSpot and RevOps partners with implementation overflow
- CRM consultants who need workflow, data, integration, or QA support
- Agencies with messy lead routing, handoff, or reporting workflows
- B2B teams with useful tools that do not communicate cleanly
- Teams that want a small fixed first step before a larger build
Usually not the fit
- Full enterprise CRM strategy from scratch
- Replacing a senior RevOps architect
- Large custom software products disguised as automation projects
- Projects with no owner, no business rules, and no source of truth
- Blind “AI agents” allowed to rewrite critical CRM data
Tools I work with
The stack depends on the workflow.
I am not selling a specific tool. I use the existing stack where it makes sense and add the minimum reliable glue around it.
CRM
Pipedrive, Salesforce, GoHighLevel, Airtable, HubSpot-adjacent workflows
Automation
Make, n8n, Zapier, native CRM workflows, REST APIs, webhooks
Data & ops
Google Sheets, Excel, Google Drive, Google Docs, Data Stores, processing ledgers
Communication
Slack, Gmail, Google Workspace, Cal.com, Fireflies
AI
OpenAI, Claude, structured extraction, classification, review-first enrichment
Other systems
Lever, Dripify, Surfe, Beehiiv, Brevo, Google Contacts, DataForSEO
Have a CRM or workflow issue your senior team should not be stuck on?
Send me the rough context. I’ll tell you honestly if I can help, and if it fits, I’ll scope a small fixed first step.