Life after Hubdoc: where document collection goes now

Hubdoc has been retired by Xero, and firms that relied on it for client document collection have three realistic paths: move to a standalone fetch-and-OCR tool like Dext, fall back to portal-and-email workflows, or adopt a platform where document intake lives inside the books themselves. Woven Financial OS, our product, takes the third approach: AI-assisted document intake, a per-client inbound email address, and portal document requests built into the ledger.

Hubdoc earned its place in a lot of firm stacks. It fetched bank statements automatically, gave every client a forwarding email, read documents passably well, and cost little because Xero bundled it. Its retirement, widely discussed in accounting communities, leaves a genuine hole. Here is an honest map of what fills it.

What did Hubdoc actually do for firms?

Worth listing precisely, because your replacement needs to cover each job or you need a plan for the gap:

1. Fetching: logging into bank and utility sites to pull statements automatically, so you did not chase clients for PDFs.
2. Inbound capture: a unique email address per client, so receipts could be forwarded and mobile photos uploaded.
3. Extraction: OCR reading vendor, date, and amount off the document.
4. Publishing: pushing the document and its data into the ledger, attached to the transaction.
5. Audit trail: the source document stored next to the entry it supports.

Most complaints about Hubdoc were about reliability of fetch connections and OCR accuracy, not the concept. The concept was right: documents should arrive without a human chasing them, and end up attached to the books.

Is Dext the natural successor?

For many firms, yes, and we will say that plainly even though we compete with the workflow. Dext (formerly Receipt Bank) is the most mature standalone product in this category. Its extraction quality is well regarded, it handles supplier statements and expense reports, it publishes to QBO and Xero, and it has firm-level dashboards for chasing missing paperwork.

The trade-offs are cost and separateness. Dext is priced as its own per-client or per-document subscription on top of your ledger spend, which is exactly the stack-sprawl problem firms complain about. And it remains a separate system: separate login, separate client list, separate place where documents live before they reach the books. If you are otherwise happy on QBO or Xero, Dext is a solid answer and you can stop reading here.

Can you just use a portal and email?

You can, and plenty of firms do. A shared folder or portal request list, a standing monthly email, and discipline. It costs nothing extra. It also reverts the work Hubdoc automated back to humans: someone chases the documents, someone renames the files, someone keys the data, and nothing is attached to transactions unless someone attaches it. It works at 15 clients and quietly breaks somewhere on the way to 50.

What does intake built into the books look like?

Disclosure: Woven is our product, and this is the approach we build.

In Woven Financial OS, document collection is not a companion app. It is a function of the ledger:

• Per-client inbound email. Each client's books get an email address. Receipts forwarded there, or dropped through the portal, land in that client's intake queue inside the platform.
• AI-assisted reading, human-verified. The AI reads the document and proposes what it is: vendor, date, amount, suggested coding, matched bank transaction where one exists. A person approves it before anything posts. Nothing is autonomous, and every client has an AI kill switch if they want document handling fully manual. During beta, AI features are included in the price.
• Portal requests. Missing the March bank statement? Issue a request from inside the engagement; the client sees it in their portal, uploads against it, and the item clears from your chase list. E-signature lives in the same portal, so the engagement letter and the document chase share one client login.
• Statement reading. Woven reads bank and card statements into transaction data directly, which matters for the clients whose banks resist feed connections.

The structural advantage is that there is no publish step and no sync to break. The document, the extracted data, the transaction, and the audit trail are the same record in the same system. The honest limitation: this only helps if you move the books themselves, and Woven is in beta with founding-firm pricing. It is a bigger decision than swapping one intake tool for another, priced accordingly at $9 to $35 per client per month wholesale.

How should a firm decide?

A practical rubric:

• Staying on QBO or Xero, need fetch and OCR now: go to Dext or a comparable standalone tool. Budget for it as a permanent per-client cost.
• Small client count, simple documents: a disciplined portal-and-email process is genuinely fine. Write the process down so it survives staff turnover.
• Already frustrated with the stack, not just the intake tool: consider whether Hubdoc's retirement is the prompt to consolidate. If you are paying separately for ledger, intake, portal, and e-signature, pricing out a one-platform alternative is rational. Woven imports chart of accounts, customers, vendors, balances, and history from QuickBooks or Xero, so a pilot with a handful of clients is a contained experiment, and the 14-day trial needs no card.

Whichever path you take, decide deliberately. The worst outcome of Hubdoc's retirement is the default one: documents drifting back to email attachments and a shared drive, and your staff quietly becoming the OCR.