How a Chicago Bankruptcy Firm's Client Intake Automations Silently Broke — and Were Fixed Before a Hard Deadline
Three Zapier workflows at a bankruptcy law firm stopped firing when Pipedrive updated its API. See how Connex identified and fixed the issue in one session.
TL;DR: A 10-person bankruptcy law firm's client intake automations silently broke when Pipedrive deprecated its stage-trigger API. Three Zapier workflows — welcome packet letters, medication mailers, and scheduling sheets — had been failing undetected, forcing staff to manually copy client data into spreadsheets. A Connex consultant identified the root cause (outdated pipeline stage IDs across all three Zaps), rebuilt the trigger logic against the new Pipedrive API, and restored full automation before a hard July 31 deprecation deadline. Estimated time recovered: 2–3 hrs/week in manual data entry.
Photo by Scott Graham on Unsplash
The Situation
A small bankruptcy law firm — 10 employees, focused entirely on consumer bankruptcy cases, based in the Chicago metro area — had been running a set of Zapier automations since 2021. The core workflow: when a potential client's deal moved to the 'Intake Scheduled' stage in Pipedrive, three automations fired in sequence, each populating a different Google Sheet with the client's name, full address, appointment date, and case type.
These sheets fed the firm's paper-based intake process — welcome packets mailed before appointments, medication questionnaire mailers for relevant case types, and scheduling confirmation letters. For a bankruptcy firm, getting the right paperwork to the right address before the first appointment is not optional. Errors here mean late clients, missed consultations, or documents going to the wrong location.
The automations had been running quietly in the background for roughly two years. Then, at some point, they stopped — and nobody noticed immediately.
The Problem
When the firm's owner reached out to Connex in June 2026, his description was direct: "Automation built by your company about two years ago, not working anymore. When a deal is moved in Pipedrive to a certain stage, the name and address should populate onto a spreadsheet — and it isn't."
The root cause turned out to be a Pipedrive API change. The Zapier 'Updated Deal Stage' trigger had been built using hardcoded pipeline stage ID numbers (integers like 39, 50, 82). Pipedrive had updated its internal stage-reference structure, making those numeric IDs stale — and the automations were silently filtering out all incoming events.
The failure mode was particularly insidious: the Zaps weren't erroring out visibly. They were simply matching zero deals because the stage filter references no longer corresponded to anything in the live Pipedrive account. Staff would occasionally notice a missing spreadsheet row and fill it in manually, not realizing the whole system had gone dark.
There was also a clock: Pipedrive had announced it was fully deprecating the legacy stage-based trigger format on July 31, 2026. Any Zap still using the old API structure would stop firing entirely after that date — not just filtering incorrectly, but completely non-functional.
What Had Already Been Tried
The firm owner had reviewed the Zapier settings himself and couldn't identify why the stage filters weren't matching.
He'd confirmed the Pipedrive stages themselves were configured correctly — the problem wasn't on the CRM side.
One of the three automations (the medication mailer) had actually continued working — which created confusion about why the other two weren't. The working Zap happened to be referencing a stage ID that was still valid, masking the pattern.
Without visibility into the internal Pipedrive stage-to-pipeline mapping, diagnosing which numeric IDs were stale versus active required CRM admin access — something the firm owner couldn't easily debug on his own.
What We Fixed
A Connex consultant joined a working session with the firm owner, got shared access to the Zapier account, and walked through all three automations systematically.
The diagnosis took less than 30 minutes:
Stage ID mismatch: The 'Welcome Packet' and 'TUIT' Zaps were filtering on pipeline stage IDs (39, 50, 82) that no longer mapped to the active 'Intakes → Intake Scheduled' stage in the live Pipedrive account. The correct stage was now pipeline 7.
Deprecated trigger: All three Zaps were using the old 'Updated Deal Stage' trigger format that Pipedrive was retiring July 31. Rebuilding them against the new API format was required regardless of the filter fix.
Missing address component: The original automations hadn't included an apartment/sub-premise field in the address mapping. Clients with apartment numbers were getting incomplete mailing addresses on their welcome packets. This had likely been causing silent errors since 2021.
The consultant rebuilt all three Zaps against the new Pipedrive trigger API, remapped the correct stage IDs by checking the active pipeline structure directly in the Pipedrive URL, and added the apartment/sub-premise field to every address step across all three workflows.
One edge case surfaced during testing: a single client record ('Jeff Daley') had an unusually long case number in the notes field that exceeded a data size limit in the note-creation automation step. The consultant confirmed with the firm owner that this was acceptable — one extreme case out of thousands — and the owner agreed to manually edit that record to prevent it. The automation ran cleanly on all other test records.
The Results
All three automations were restored and validated before the July 31 Pipedrive deprecation deadline. The firm's welcome packet, medication mailer, and TUIT sheet workflows are now running on the new Pipedrive API structure and will continue operating past the deprecation date.
Key outcomes:
Automations restored: 3 of 3 Zapier workflows re-activated, now running against the correct, non-deprecated Pipedrive API
Address quality improved: Apartment/suite field added to all address mappings — a data gap that had existed since the original 2021 build
Manual entry eliminated (estimated): At ~20–30 intakes/month and approximately 5–8 minutes of manual data entry per intake when automations were down, restoring all three workflows recovers an estimated 2–3 hours/week in administrative time
Deprecation risk removed: Without the fix, all three automations would have permanently stopped firing after July 31, 2026 — requiring a full rebuild at higher cost and likely causing a gap in client communications
Non-numeric note: For a bankruptcy practice, the downstream value of reliable address data goes beyond staff time. Welcome packets reaching the wrong address or arriving without apartment numbers means clients don't have their paperwork before their first consultation — which creates rescheduling cycles, frustrated clients, and avoidable front-desk load.
What Made This Fixable in One Session
🔧
Zapier automations built against numeric pipeline stage IDs are fragile by design. Pipedrive's internal stage numbers change when pipelines are reorganized, renamed, or migrated — and Zapier doesn't surface these mismatches as errors. The Zap keeps running; it just matches nothing. Rebuilding against stage names (not IDs) and using the current Pipedrive API trigger format is the right long-term pattern.
The firm owner had originally handed the automation work to Connex in 2021 because he'd gotten close on his own but kept hitting walls. When the automation broke years later, the same barrier applied: the fix required knowing where to look (Pipedrive URL → pipeline stage IDs), having the right Zapier access, and understanding the API deprecation timeline — none of which are obvious to someone who isn't in these systems regularly.
The re-engagement fit Connex's ZoomFlow model: a focused working session where the consultant is already in the client's accounts, diagnosing and fixing live, not just advising. Total session time was under an hour.
Frequently Asked Questions
Our automations were 'working' for years. Why would they just break?
Zapier Zaps using Pipedrive's 'Updated Deal Stage' trigger reference internal stage ID numbers. When a Pipedrive account is reorganized — pipelines renamed, stages moved, IDs reassigned — the Zap's filter references go stale. The Zap doesn't error; it just matches zero deals and does nothing. This can go undetected for months if the failure is gradual (fewer rows appearing in a spreadsheet rather than a complete stop). The July 2026 Pipedrive deprecation made it impossible to ignore.
How do we prevent this from happening again?
Two practices help: (1) Reference stage names rather than numeric IDs where possible, and test automations any time a Pipedrive pipeline is reorganized. (2) Build a lightweight monitoring step — even a simple Slack or email notification that fires when a Zap processes zero records in a 48-hour window — to catch silent failures before they compound.
We have multiple automations we're not sure are still working. What's the process?
The fastest approach is a Zapier audit session: get a Connex consultant into your Zapier account alongside your CRM, run test deals through each workflow, and confirm which Zaps are processing records correctly versus filtering everything out. For most small practices, this takes 1–2 sessions. Contact us to schedule.
Is your intake automation quietly broken?
If your firm has Zapier automations connected to Pipedrive that were built more than a year ago — especially any that trigger on deal stage changes — they may be silently failing. The Pipedrive API deprecation on July 31, 2026 will make this impossible to ignore. Book a free ZoomFlow session to check and fix your automations before the deadline.
How a regional mortgage lender automated offline conversion tracking across Google, Bing, and Meta using Zapier and the Velocify API — no weekly manual exports.
How a mid-sized countertop manufacturer replaced their inbound status-inquiry phone queue with an AI voice agent connected to Moraware. Read the case study.
