Integrations
Fiserv integration for financial spreading
How a finished SpreadSpace spread serves a bank on a Fiserv core: an export map in its codes, file and API handoffs, and Fiserv’s Banking Hub for core data.
Fiserv runs the core for a large share of American banks and credit unions, on Premier, Signature, DNA, Finxact and its other platforms, and its Developer Studio opens those cores to outside software through the Banking Hub APIs. Commercial loan origination at a Fiserv institution usually runs in a separate loan origination system that boards the finished loan onto the core. A lender that reviews borrowers in SpreadSpace needs the finished spread in that system, and sometimes the core’s own records in the spread.
This guide covers both directions as they work today. SpreadSpace names every account in the terms the lender’s origination system expects through a per-workspace export map, hands the finished spread over as a file, through the API when the analyst finalizes, or inside an embedded workspace, and the lender’s team reads core data through Fiserv’s documented APIs. No Fiserv-built connector exists, and none of the routes needs one.
Where a spread lands at a Fiserv institution
Fiserv’s Banking Hub, inside its Developer Studio, gives developers one place to reach its core banking APIs. Fiserv launched it with development environments for Premier and Signature, with Finxact alongside and DNA to follow, with self-service workspaces and trial accounts. Those APIs read and write the customer, account and loan records the core holds. Spreading and credit analysis are not core functions, so they live in the origination system the institution runs beside the core.
A spread finished in SpreadSpace lands in that origination system, in the accounts its templates expect, and the loan reaches the Fiserv core through the system’s boarding. The core comes back into the spread when the analyst needs the borrower’s existing accounts and loans at the bank, which the lender’s team reads through the Banking Hub.
Fiserv on the Banking Hub and Developer Studio (2023 release)
What SpreadSpace hands to Fiserv
Four things leave SpreadSpace. The first is the extracted document: every line a tax return, financial statement or bank statement prints, each with the page and the boxed region it was read from. The second is the mapped statement: the profit and loss statement and the balance sheet placed on the workspace’s chart of accounts, with each account’s figure per period and the arithmetic that walks the printed lines to the printed totals. The third is the finalized spread: the attribute snapshot an analyst approved, with its figures, periods and source document references, frozen at a version number. The fourth is the credit memo as a PDF.
Each has a file form and an API form. One document exports as CSV or JSON on its own, and a set of documents on one loan bundles in the background as CSV, JSON or, for bank statements, XLSX. The loan’s spread comes out as two Excel workbooks, one of statements and one of tables, with the rows a figure is derived from indented under it. The finalized figures are read from the loan’s attributes endpoint and the mapped statements from the borrower’s mapping endpoints. Fiserv receives whichever form its import takes.
Exports in the API documentation
Name every account in Fiserv’s terms
The chart of accounts is the workspace’s own. Its accounts carry SpreadSpace ids such as pl.opex.rent and bs.current_assets.cash, and Fiserv holds the same accounts under its own codes and names. The export map is the translation. In Settings, open Chart of accounts and then Export map, choose Fiserv as the system, and give each account the code and the name Fiserv uses for it. The map is saved once per workspace and applies to every loan, and the system is recorded as fiserv on everything it codes.
| SpreadSpace account | Account id | Code and name in Fiserv |
|---|---|---|
| Revenue | pl.revenue.sales | 4000 Sales |
| Officer compensation | pl.opex.officer | 6100 Officer compensation |
| Rent | pl.opex.rent | 6300 Rent |
| Cash | bs.current_assets.cash | 1000 Cash and due from banks |
| Accounts receivable | bs.current_assets.accounts_receivable | 1200 Accounts receivable |
| Lines of credit | bs.current_liabilities.lines_of_credit | 2200 Revolving line |
| Retained earnings | bs.equity.retained_earnings | 3900 Retained earnings |
From then on every mapped statement carries the codes. Each account in the mapping’s tree answers with its external code and label beside its figure per period, the payload names the system, and the accounts the map codes that the statement does not use are listed, so nothing posts blind. A map can also be saved through the API, up to 1,000 entries and the whole map each time, and a save that crosses another is refused rather than merged. After any save, each mapping reads as stale until it has been prepared again with the new codes, and a webhook announces the fresh one.
Statement mapping and the export map in the API documentation
Three routes into Fiserv
The first route is a file. An analyst downloads a document’s export or a bundle from the document list, a job requests the two workbooks, and the files are loaded through Fiserv’s import or attached to the loan. It needs no engineering and suits a team that files a few spreads a day.
The second route is the API. When an analyst finalizes a spread, SpreadSpace sends a spread.finalized event to your endpoint, signed with a shared secret, naming the loan and the snapshot. Your service reads the snapshot from the loan’s attributes endpoint and the mapped statements from the borrower’s mapping endpoints, with Fiserv’s codes already on every account, and posts them into Fiserv through the path Fiserv documents. SDKs for TypeScript, Python and C# wrap the calls, and the event carries the snapshot id, so a handler that records the ids it has posted stays idempotent under a retry.
The third route is the embedded workspace. Your application asks SpreadSpace for a single-use signed URL and seats the spread in an iframe beside the Fiserv record, so an analyst opens the source documents without leaving the loan, and no token crosses the parent frame. The figures still finalize into the same snapshot and reach Fiserv by either route above.
Webhooks in the API documentation
Fiserv’s side of the handoff
Fiserv’s documented path for outside software is the Developer Studio: a workspace, an API key and the core platform’s API set. That is the route for reading the core, and for writing to it only where the institution’s Fiserv arrangement allows. The finished spread itself follows the origination system’s import, which this site covers for each of the common ones, from nCino and Abrigo to Baker Hill and Jack Henry.
On the API route your service receives the finalize event, reads the snapshot and the mapped statements with the origination system’s codes on every account, and posts them there, with any core lookups made through the Banking Hub on the way. On the file route the analyst attaches the statements workbook or the CSV bundle to the credit file. The embedded workspace keeps the source documents beside the record in either case.
Set up the handoff
Create an API key with the scopes the route needs: extractions read and export for the files and the mapped statements, spreads read for the snapshot, chart of accounts write to save the map from code, webhooks write to register the endpoint and embed write for the iframe. Save the export map for Fiserv. Register the webhook endpoint, send the test event and verify its signature. Then run the whole loop on a test key against the sandbox loan, whose sample spread finalizes on request, before the first live loan.
SpreadSpace ships no connector written by Fiserv, and no step above depends on one. The file, the API payload and the embedded frame are SpreadSpace’s side of the handoff. Fiserv’s side is its documented import, its API or its services team, and the section above names the public pages that describe it. A lender that wants the two sides joined for it can ask sales.
API keys and scopes in the documentation
Sample documents and the sandbox loan in the documentation
Integrating financial spreading with a loan origination system