Direct Answer: Treat Cap-Table Migration as a Transaction, Not a File Transfer
A startup cap-table migration is the controlled transfer of equity ownership, shareholder records, grant history, option records, tax information, and supporting documents from one system or service provider to another. Founders responding to a shutdown, changing banks, consolidating software vendors, or entering a new financing should begin immediately rather than waiting for an export folder. The immediate objective is continuity: investors, employees, option holders, and service providers must be able to verify who owns what as of a specific date.
Also worth reading: How Does AI Investor Matching Work for Startup Founders in 2026? · How Should AI Startup Founders Plan for Dilution Before Raising Their Next Round? · How Is AI Transforming Startup Deal Flow for Founders and Operators?
The research context points to Pulley’s announced shutdown on December 8, with Carta reportedly absorbing or serving customers afterward. That creates a concrete reason for affected U.S. startups to verify whether they have received migration instructions, rather than assume that customer transfer is automatic. A cap table is not merely a spreadsheet because it can determine voting power, liquidation proceeds, 409A valuations, option grants, tax reporting, and the company’s ability to raise capital. A poorly reconciled migration may therefore create legal, accounting, payroll, and financing problems even when the company has changed nothing about its operations.
The safest process is to export the complete data, establish a cut-off date, reconcile every security and holder, import the records into the destination, test the results against the old system, obtain professional review where needed, and retain an audit trail. Companies should not treat a vendor’s statement that the transfer is “complete” as proof that every historical transaction, payment, cancellation, or beneficial ownership detail is accurate. Responsibility for the company’s records ultimately remains inside the company; service migration does not transfer that responsibility away from founders and counsel.
Why a Cap-Table Migration Becomes More Complicated Than It Looks
Cap tables often contain several record types that must remain synchronized: common and preferred shares, options, warrants, convertible securities, SAFEs or other promissory instruments, transfers, cancellations, exercise prices, vesting schedules, and historical grants. Each record can have multiple owners or beneficiaries, and a company may maintain different versions in its board minutes, stock ledger, general ledger, payroll provider, and legal records. Migration software can move structured data effectively, but it may not interpret an ambiguous historical entry or reconcile conflicting source documents.
A migration also affects more than the spreadsheet. The company must preserve evidence supporting federal and state tax filings, including Form 1099-B and Form 1099-DISC where applicable, while maintaining the records required for option grants, payroll tax withholding, securities compliance, and corporate governance. In a private company, international holders and cross-border grants can add withholding, foreign exchange, and local securities-law questions. A destination platform may be perfectly capable of storing the basic share count while still lacking a required tax workflow or document for a particular jurisdiction.
Timing is another source of risk. Founders sometimes begin planning after a shutdown notice, then discover that data exports expire, administrator access is restricted, or employees cannot retrieve historical grant confirmations. This is why September 29, 2026 is a meaningful checkpoint for companies affected by the reported December 8 Pulley shutdown: they have roughly ten weeks to export, reconcile, test, and communicate, assuming the published shutdown date and migration schedule remain unchanged. Companies should treat any deadline as potentially earlier than the final shutdown because support volume can rise near the date, and they should avoid making irreversible changes during the testing stage.
The Step-by-Step Migration Method for a Startup
First, appoint one migration owner and create a source-of-truth inventory. The owner should identify the current legal cap table, transaction ledger, board-approved grants, option-plan records, tax reports, bank or payment records, and all relevant historical spreadsheets. The inventory should state which system is authoritative when records conflict, who holds administrator access, and whether a service contract restricts export. Founders should preserve read-only access wherever possible so that the source cannot change while the destination is being built.
Second, freeze or control changes to equity records. This does not necessarily mean hiring or paying nobody for 70 days; it means preventing untracked grants, exercises, cancellations, transfers, or plan amendments. If a new grant or exercise must occur, it should be entered in the current system and the migration log with an exact effective date. Parallel bookkeeping can be necessary during the transition, but each transaction must have one designated home and a reconciliation method.
Third, export the broadest available dataset. That should include securities, holders, transactions, grants, vesting schedules, documents, tax information, plan versions, and audit logs where available. CSV or spreadsheet files are useful for inspection, but a native export or direct transfer can preserve field types and relationships more reliably. Founders should also request the data dictionary, encoding details, and export documentation, because dates, currency values, security classes, and identifiers can be reformatted during conversion.
Fourth, reconcile totals before importing. At minimum, compare issued and outstanding shares by class, options granted, options exercised, options available, warrants, convertible instruments, and fully diluted totals. The migration team should sample every class and several ordinary transactions rather than rely only on a top-line total. A matching total can conceal errors such as two grants loaded against the same participant, an incorrect vesting start date, or a canceled security reappearing as outstanding.
Testing the Destination System and Preserving an Audit Trail
Testing should occur in a nonproduction environment or a clearly marked duplicate of the company’s records. The test should begin with a small, representative set that includes common shares, each class of preferred stock, an option grant, a vesting schedule, an exercise or cancellation where applicable, and at least one multi-class or multi-holder relationship. After loading the sample, compare identifiers, ownership quantities, ownership percentages, issue dates, vesting dates, payment status, and document links between source and destination.
The team should also run business-level checks. Fully diluted ownership should remain unchanged unless a known transaction is intentionally excluded, and the sum of issued shares should agree with the stock ledger and board records. Historical option balances should distinguish authorized, granted, vested, exercised, canceled, and remaining available shares. These tests matter because migration errors often appear not in share count but in a subsidiary calculation that affects an employee’s grant, an investor’s return model, or the next financing’s capitalization table.
A migration audit trail should include the source-export date, destination-import date, file hashes when practical, administrator actions, transformation rules, exception decisions, and approval from the responsible officer, accountant, or counsel. Keep both the original export and the final import evidence for at least as long as tax, securities, employment, and contract requirements advise. The company should not delete the prior system until at least one successful reporting cycle and an independent review are complete. Pulley’s reported shutdown illustrates why a backup matters, but a backup without a known recovery procedure has limited value.
Comparing Major Migration Routes
There is no universally best option. A small company with a clean, uncomplicated ledger may manage a spreadsheet migration with professional review, while a venture-backed company usually needs institutional-grade records, permissions, tax workflows, and an experienced administrator. The destination vendor matters less if the company cannot access the underlying data, understand the conversion, or produce reliable reports afterward.
| Feature | Spreadsheet or self-managed migration | Dedicated cap-table platform migration |
|---|---|---|
| Best fit | Simple cap tables, low transaction volume, limited budget | Companies with investors, employees, grants, or frequent financing |
| Typical software cost | $0 to several hundred dollars in tools, plus labor | Often roughly $100 to $2,000+ per year for standard plans; enterprise pricing may be higher |
| Main strength | Flexible, familiar, and inexpensive | Centralized records, permissions, reporting, and audit features |
| Main weakness | Human error, weak access controls, and poor change tracking | Subscription cost, vendor dependence, and migration configuration work |
| Suitable migration test | Independent reconciliation of every class and security | Parallel report comparison, exception log, and administrator sign-off |
| Long-term option | Acceptable only with strong controls | Usually preferable for a growing startup |
When Founders Should Act Immediately
A shutdown deadline should trigger immediate action. Because the reported Pulley shutdown date is December 8, a company waiting until November could face compressed support queues, restricted access, or an expired export window. The first 72 hours should be used to secure administrator credentials, export current data, archive documents, inventory integrations, and identify unresolved discrepancies. Founders should confirm the exact date and scope of Carta’s reported customer transition directly with both providers, because reporting can simplify a process that has separate deadlines for data access, billing, service, and support.
A planned migration should begin earlier when the company has a closing financing, an upcoming board consent, a 409A refresh, a stock sale, or a payroll or tax-reporting dependency. Companies with fewer than 25 stakeholders and no unusual instruments may finish a basic reconciliation within a few business days, but estimates should be made only after reviewing data quality. A venture-backed company with multiple preferred classes, a large option pool, international participants, and years of amendments may need several weeks of administrator work and external review.
Several thresholds can serve as practical warning signals, although they are not legal tests. A company with more than 50 stakeholders, more than five security classes, ongoing monthly grants, or an upcoming institutional financing is unlikely to be well served by an improvised spreadsheet. Any discrepancy in authorized shares, issued shares, or plan reserves should stop the migration until resolved. Likewise, a company that cannot identify who approved a historical grant should not simply copy the entry forward without a documented review.
Common Mistakes and How to Avoid Them
The most common mistake is waiting for an email announcing that migration is complete. Founders should independently compare the old and new reports before allowing the new system to become the official record. Another error is treating ownership percentages as equivalent to security counts; percentages can change with the cap-table denominator, while share quantities should remain stable except for intentional transactions. Companies also err when they overlook authorized but unissued options, warrants, or converted instruments.
A second major mistake is mixing actual ownership with future potential. A migration must preserve granted, vested, exercised, canceled, and available options rather than compressing them into one number. Founders should verify that grant dates, vesting commencement dates, cliff rules, acceleration provisions, and exercise prices survived the transfer. Tax documents, board approvals, and shareholder agreements are equally important because a data record without its supporting evidence can be difficult to defend during diligence.
The final mistake is failing to communicate. Employees and investors should receive a clear notice explaining whether their ownership, grant, tax reporting, or payment arrangements are changing, along with the action they need to take and the date by which it must occur. A migration should not be used to quietly change economic terms, vesting, or reporting obligations. If the destination provider lacks a necessary function, the company should identify a temporary manual process and assign responsibility rather than assume the issue will disappear when the migration ends.
A Practical Ownership and Decision Framework for Founders
The decision to migrate should be based on operational risk, cost, and future control rather than brand loyalty. A platform with stronger permissions and reporting may be worthwhile even if it costs more, especially when the cap table influences financing or employee compensation. A simpler system can be sufficient for a very small company, but only if the founder understands how to maintain it and obtains help when a transaction is unusual. The right choice is the one that produces accurate, explainable records under the company’s actual circumstances.
The Mercer Club audience, which includes founders and operators, should view the migration as a short data-governance project. The site can offer a neutral workflow, questions to ask providers, and a way to compare alternatives without implying that any software or AI network automatically solves cap-table risk. AI can help classify transactions, identify duplicate holders, draft exception summaries, or compare exports, but it should not independently decide legal ownership, tax treatment, or the validity of an equity grant. Automated review still needs deterministic reconciliation and accountable human approval.
For companies affected by a vendor shutdown, the most important sequence is simple: preserve access, export broadly, reconcile precisely, test in a duplicate, obtain approval, communicate, and retain evidence. The research references also suggest a broader market transition in cap-table management, including reported movement of Pulley customers toward Carta, but founders should not substitute industry commentary for contractual notices or official migration instructions. The company’s own records, governing documents, tax obligations, and professional advisers should determine the final result.