Hours saved annually
Annual cost reduction
Error rate reduction
Faster invoice processing
We were drowning in invoices. Every acquisition brought a new system, a new approval chain, and more manual work. We needed a solution that could grow with us—not slow us down.
– Chief Technology Officer
Staff hours reclaimed from manual data entry and rework.
Error rate reduction across all 14 entities
Invoice reconciliation time at month-end close
Annual savings from early-payment discount capture.
Finance controllers across all 14 entities reported measurable improvement in month-end close timelines. Four FTEs previously dedicated exclusively to invoice data entry were reassigned to higher-value activities — vendor relationship management, cash flow forecasting, and financial analysis — improving both employee satisfaction and strategic output.
The platform’s modular architecture means each future acquisition can be onboarded to the automated AP pipeline in under two weeks, with no additional headcount. The organisation has already absorbed three new agency acquisitions since go-live without any increase in AP staffing.
Reduced manual effort in invoice processing
Faster processing cycle time end-to-end
Improved accuracy in data extraction and mapping
Eliminated dependency on manual media team communication
Scalable across multiple agencies and suppliers
Within 90 days of full go-live, the AP team had reclaimed an estimated 810 staff hours per month that were previously lost to manual data entry and rework. Finance controllers across all 14 entities reported a measurable improvement in month-end close timelines, with invoice-related reconciliation tasks reduced from an average of 4.2 days to under 18 hours. The error rate dropped from approximately 12% to under 0.6%, with all exceptions automatically flagged and routed — eliminating the need for manual invoice audits.
Staff redeployment was a significant secondary benefit. Four FTEs previously dedicated exclusively to invoice data entry were reassigned to higher-value activities, including vendor relationship management, cash flow forecasting, and financial analysis — improving both employee satisfaction and the strategic output of the finance function.
This isn’t just automation — it’s an entirely different way of running AP. We processed more invoices last quarter than ever before, with fewer people and zero escalations to the CFO. The ROI was visible within the first month.”
– Chief Technology Officer
When marketing networks grow through acquisition, each entity brings its own AP workflow, Freshdesk queue, and supplier relationships. Automating financial consolidation at the invoice level — using LangGraph agents to coordinate data across Prisma and Maconomy — eliminates compounding errors and ensures accuracy at enterprise scale.
Rules-based RPA and traditional OCR tooling cannot handle the real-world variability of vendor invoices across formats, suppliers, and currencies. Gemini 2.5 Flash's native document intelligence capabilities enabled reliable extraction from structured and unstructured invoices alike — without a separate OCR layer — reducing exception handling by 95%.
The most significant bottleneck was not invoice entry itself but the dependency on media team chat for Order and Placement IDs. LangGraph agents addressed this root cause by automatically extracting these identifiers from communications — removing the follow-up cycle that was the primary source of delays and errors.
The network successfully onboarded three new agency acquisitions post-deployment without adding a single AP staff member. The automated pipeline absorbed the increased invoice volume seamlessly – demonstrating how intelligent automation future-proofs finance operations against both organic and inorganic growth.
Follow Us
