The Results
| Metric | Result |
|---|---|
| Holders processed | ~1,400 across shareholders, warrants, options, COC, and LOAN records |
| Time saved | Weeks of manual reconciliation compressed into hours per rebuild cycle |
| Deadline | TSX-filed circular delivered on time |
| Data reach | Addresses reconciled across 21 international jurisdictions against an external geocoder |
| Deliverables | Consolidated mailing workbook, TSX-formatted multi-tab workbook, shareholder database |
The Challenge
A TSX-listed issuer was preparing a shareholder circular for an M&A transaction. The filing — and the mailing behind it — had to be right, and it had to be on time. Missed ballots on a TSX circular are not a fix-it-tomorrow problem.
The underlying data lived across three unrelated sources: a registry export, an internal cap-table workbook, and a scattered email-address spreadsheet. Addresses were inconsistent across twenty-one jurisdictions, some holders appeared in multiple forms, and a handful required two envelopes by explicit client direction. None of it could be trusted for a regulated mailing without a full reconciliation — and there wasn't time to do that by hand.
The Solution
I was brought in to consolidate the data and deliver a circular the company could file with confidence. The approach was full-stack: clean the inputs, build the system, ship the artifacts.
Address cleanup pipeline. Fuzzy city matcher and jurisdiction-aware fix layer, audited against corrected geocoder output.
Consolidated mailing list. One row per holder, with explicit multi-envelope exceptions flagged for review.
TSX-formatted multi-tab deliverable. A five-tab workbook (Shareholders, Warrants, Options, COC, LOAN) versioned through two dozen iterations.
Shareholder database. A working register of the full shareholder base, delivered alongside the circular.
All driven by Python and Excel automation that preserved the quirks of the real workbooks (hidden sheets, formula-bearing cells, formatting the client relied on) rather than flattening them into something cleaner but wrong.
Key Takeaway
This is the shape of a lot of business operations: the data exists, but it lives across three systems that were never designed to talk to each other, and when a deadline hits, someone has to reconcile it by hand. That works until it doesn't — until the stakes are a regulated filing, the volume is 1,400 holders across 21 jurisdictions, and there's no time left. Clarys's job in a project like this is to understand the operation end-to-end, build the pipeline that does the reconciliation reliably, and hand the client artifacts they can sign their name to. The circular filed on time, the workbooks match, the mailing delivered.