Datarails built a future for finance by refusing to fire Excel

At Cisco, Didi Gurfinkel expected the end of the month to look like the future: press a button, watch the numbers arrive. Instead, spreadsheets travelled between his business unit and California, accompanied by plenty of manual work. A technology company could connect the world, yet its own financial information still needed a chaperone. That experience supplied the question behind Datarails: could the spreadsheet become an enterprise tool without ceasing to be a spreadsheet?

Four years to find the right desk

Gurfinkel founded Datarails with Eyal Cohen and Oded Har-Tal in 2015. The original ambition was broad: connect organizational spreadsheets to a central database. In his later account, finding product-market fit took four years. Finance supplied the focus. Its recurring need to collect differently shaped data made consolidation a business problem with a recognizable owner and a deadline.

The lesson travels beyond accounting. A useful technology can remain commercially vague until someone needs it every month. Datarails found that someone in the CFO’s office, where the familiar workbook carried years of accumulated judgment. Asking a team to abandon it would mean asking people to rebuild part of their working memory.

The trouble starts between the cells

Datarails sells cloud software for financial planning and analysis, or FP&A. Teams use it to consolidate figures, build budgets, update forecasts, compare actual results with plans and produce financial reports. Excel remains a working surface. The infrastructure behind it collects and organizes data that would otherwise arrive as disconnected exports.

The integration catalogue includes NetSuite, QuickBooks, Sage Intacct, Salesforce and payroll systems. Those connections matter because a finance question rarely respects software boundaries. Hiring plans live near HR; sales expectations live near the CRM; actual expenditure lives in accounting. A forecast needs those departments to tell a compatible story.

The documented Flex add-in makes the mechanics less mysterious. Refresh brings current Datarails data into an Excel report. Submit saves a new file version rather than overwriting the previous one, recording when it was created and by whom. Mundane buttons, useful consequences: fewer disputes over which workbook deserves belief.

A budget is an excellent place to lose a week

At NovaTech Automation, the old budget travelled by email. Department managers returned spreadsheets; finance typed their figures into a central database. Revisions could send the exercise back around. The initial search was for easier budgeting, but better KPI reporting widened the brief. The purchase decision changed when budgeting became the entry point to a broader view of operations.

In a Datarails-published case study, NovaTech reports saving roughly 80 hours on annual budgeting and six to eight hours on monthly reporting. Better visibility also helped pricing decisions during inflation. These are customer-reported outcomes, rather than a promise that the next buyer will save the same amount.

“Now there’s a single version of what’s really happening.”BJ Weil · NovaTech Automation

The practical thing to copy is the sequence: identify repeated entry, connect the underlying records, then ask better questions. Faster preparation gives finance more time to inspect margins, delivery performance and assumptions. Automating a bad definition of revenue would merely distribute confusion more efficiently.

The price of keeping the familiar

The customers are CFOs, controllers and FP&A teams, particularly spreadsheet-centered small and mid-market businesses. Datarails’ current website claims adoption by finance teams at more than 2,000 companies across industries including manufacturing, healthcare and property management. The appeal rests on recurring complexity, rather than belonging to one fashionable sector.

Its demo form says plans begin at $1,000 per month. Exact quotes depend on scope. The published packages scale users and integrations, with additional finance modules at higher tiers. Buyers should distinguish this packaged software pricing from the usage-based model announced for standalone FinanceOS.

Excel connectivity alone cannot settle the competitive argument. Cube also connects spreadsheets to a governed model. Anaplan offers connected financial planning and scenario modeling. Datarails’ particular pitch combines retained Excel workflows, financial consolidation and an expanding finance suite. The sensible comparison is a demonstration using your own awkward report.

Growth met its own budget meeting

The company’s history has a less cheerful column. In December 2022, Datarails cut 30 jobs, mostly in US sales and technical support. Management said it had hired for rapid growth, then shifted its goal toward profitability as market expectations changed. A financial planning company had to revise its own plan.

By January 2026, the company reported 70% revenue growth during 2025 and more than 400 employees. One Peak led a $70 million Series C. The contrast is instructive: demand, staffing and spending do not automatically move at the same sensible speed.

Now the spreadsheet gets company

With FinanceOS’ March 2026 introduction, Datarails made its underlying data infrastructure the pitch. The bet: customers will want to use outside AI tools, while finance still needs permissions, reconciled numbers and an audit trail. Its AI documentation describes read-only access that respects existing permissions and logs queries.

The new services also put people back into the software story. Public US recruiting materials describe cross-functional support work, remote roles and professional development. Those are employer claims, but they point to a practical requirement: financial automation needs people who can explain both the product and the accounting problem.

The territory keeps widening. August brought embedded financial engineers to build custom AI workflows. September brought Desk, which links finance requests to supporting records and drafts replies for human review. Cash management, close and spend control extend the same foundation into adjacent work.

This approach makes sense when Excel is valuable and repeated consolidation is expensive. It becomes harder to justify when reporting is simple, existing systems already handle the work, or nobody owns the data definitions. AI still needs someone to judge assumptions. The spreadsheet has survived; the interesting question is how much clerical baggage it must continue to carry.