Home
Knowledge
Cash Forecasting
Cash Forecasting - Current Solutions & Workarounds
What it is: Gut feel and experience-based trust in forecasts
How they use it: Treasury teams develop intuition about when forecasts are reliable
Limitations: Not explainable to stakeholders, not systematic
Used by: ON
Peer Validation
What it is: Talk to other treasury teams (e.g., HelloFresh) about forecast reliability
How they use it: Shared experience builds confidence when no formal metrics exist
Limitations: Anecdotal, not data-driven
Source: ON (2025-11-11)
Conservative Assumptions
What it is: Apply manual adjustments or use conservative scenarios when trust is low
How they use it: Can't fully rely on model output without understanding it
Source: ON (2025-11-11)
Excel
What it is: Spreadsheet software
How they use it: Primary forecasting tool, manual aggregation of entity data
Limitations: Manual, no real-time updates, difficult to maintain across entities
Used by: Euroports
Cash Analytics
What it is: Treasury Management System
How they use it: Cash visibility, but manual input only (no ERP integration)
Limitations: Not designed for forecasting, limited analytical capabilities
Used by: Euroports
ERP Systems
What it is: Enterprise resource planning systems
How they use it: Source of AR/AP data for forecast inputs
Limitations: Only updated monthly, not useful for real-time forecasting
Used by: Euroports
SAP
What it is: Enterprise resource planning system
How they use it: Common ERP across ~40 entities; cash management module built in 2016; source of AP/AR data
Limitations: Not designed for treasury forecasting; need to extract data by due date range
Key insight: Standardization seen as critical - "I've seen big companies try to do forecasting with six or seven different ERPs and I don't think it really works so well"
Used by: Volvo Cars
Quantum
What it is: Treasury Management System
How they use it: Entities upload Excel forecasts; central Treasury uses for cash positioning and weekly/monthly forecasts
Limitations: Relies on manual Excel uploads; overwriting forecasts loses variance history
Used by: Volvo Cars
TIS Payments + Integrity TMS
What it is: Payment hub (TIS) + Treasury Management System (Integrity), connected via SFTP
How they use it: TIS for bank connectivity (fast setup, ~1 bank/month); Integrity for target balancing and cash pooling logic
Limitations: Requires manual payment run requests; not fully automated zero-balancing (since different banks)
Key insight: TMS-based pooling as interim solution when traditional bank cash pool not feasible
Used by: Live Events (2025-04-01)
Kyriba
What it is: Treasury Management System with cash forecasting module
How they use it: Open AR + Open AP + manual forecasts (salaries, taxes, significant payments) = forecast
Limitations:
"Nothing super smart, nothing of AI" - just open items and manual inputs
No category-level forecast views (can't see development of salaries, Capex separately)
No variance analysis capability ("Variance is not there. No comparison.")
Cash pool accounts show misleading negative balances
Forecasting module can't import external data - cash flows must be generated within Kyriba (IHG)
Cash Position Worksheets limited: single currency only, not customizable, changes disappear (IHG)
Liquidity Planning module (can import) costs extra (IHG)
Module inconsistency: different capabilities in different modules (IHG)
Used by: ON, IHG
HighRadius
What it is: AR/Collections management platform
How they use it: Customer payment pattern recognition, "cash application" (matching payments to invoices)
Limitations:
Customer behavior data doesn't flow to cash forecasting
Knows customer patterns (e.g., "Foot Locker always pays on the 20th") but this intelligence is siloed
Used by: ON
Dynamics
What it is: Microsoft ERP system
How they use it: Source of open AR/AP data for forecasts
Limitations:
Timing issues: paid items still show as open until accounting closes them
Causes duplicate forecasts when items are paid but not yet settled
Used by: ON
Manual Workarounds
Entity-by-Entity Excel Collection
What they do: Collect forecasts from each entity via email/spreadsheets, manually consolidate
Why: No centralized system supports decentralized input with group consolidation
Sources: Euroports (2025-10-27), Volvo Cars (2025-03-27), IHG (2025-04-07) - India service center collates spreadsheets
Offshore Service Center for Forecast Collation
What they do: Shared service center (e.g., India) sends spreadsheets of subsidiary cash flows; Treasury dealer collates into master file
Why: Centralized Treasury team doesn't have direct access to subsidiary systems; need human verification
Limitations: Manual, prone to errors (payroll runs missed), time-consuming
Source: IHG (2025-04-07) - "Sometimes the guys in India will just miss, they'll just forget to add something... even like a payroll run has been missed off"
Day Sweeps for Cash Pooling
What they do: Automatic daily sweeps; entities don't hold cash; central Treasury covers negative balances
Why: Centralized cash management; entities focus on forecasting not funding
Source: Volvo Cars (2025-03-27)
Annual Budget as Forecast Baseline
What they do: Use annual budget as starting point, make manual adjustments
Why: Only 1-2 formal forecast cycles per year
Source: Euroports (2025-10-27)
Kyriba + Google Sheets Split
What they do: Use Kyriba for actuals only; all forecasting in Google Sheets with quarterly version saves
Why: Kyriba lacks forecasting capabilities; need to track forecast accuracy over time
Source: Sonder (2024-10-03) - "We don't do any forecasting in Kyriba really"
Quarterly Forecast Version Saving
What they do: Save forecast versions each quarter; compare original forecasts to actuals over time
Why: Need to measure forecast accuracy and learn from variances
Source: Sonder (2024-10-03) - "We can go back and say, well, how accurate were we forecasting the end of October at the start of October?"
What they do: Simple forecast formula: starting balance + AP data from ERP + manual inputs (tax, salary estimates from stakeholders)
Why: Foundation of trustworthy short-term forecast based on booked data
Source: ON (2024-11-19)
Conservative Buffer Calculation
What they do: Calculate buffer based on historical variance; more unforeseen events = bigger buffer; "be pessimistic about cash-ins"
Why: You have to pay on time but won't receive on time; need cushion for forecast uncertainty
Source: ON (2024-11-19)
13-Week Rolling Daily Forecast
What they do: Maintain rolling 13-week forecast updated daily in ~15 minutes; actuals replace forecasts with variance analysis
Why: Standard process for daily cash management; balance efficiency with accuracy
Source: Sonder (2024-11-19)
What they do: Different touchpoints per team - calls with real estate, Google Sheets with tax, automated reports from payroll
Why: Different teams have different data availability and reliability levels; adapt collection method to each
Source: Sonder (2024-11-19)
Conservative Inflow Forecasting
What they do: Forecast lower on collections; "if you sign 8, put 6" - upside surprises are better than shortfalls
Why: Cash shortfalls cause operational problems; excess cash is easily deployed
Source: Sonder (2024-11-19) - "For cash flows, if you want to be conservative on the inflow... You'd rather miss a little too low than too high"
Industry Competitive Landscape (Treasury Dragons, 2025-12-09)
TIS
Positioning: Enterprise multinationals with geographic/entity complexity
Approach: Data integration first ("walk, run, fly" implementation), Working Capital Insights module (DSO/DPO)
AI use: Customer payment pattern recognition, natural language prompts for configuration, agentic AI planned for 2026
Differentiator: Strong entity collaboration workflows, two-way communication with subsidiaries
Cobase
Positioning: Bank connectivity first (origin as multibank payment platform)
Approach: Connect all banks and data sources → build treasury modules on top
AI use: Transaction categorization
Differentiator: Native multibank platform, workflow/approval flows for central treasury control
Panax
Positioning: AI native for mid-market (complex needs, lean teams)
Approach: Fast implementation (3-8 weeks), cleaning fragmented data
AI use: AI in every layer - data monitoring, categorization, ERP mapping, LLM-powered insights chat
Differentiator: Speed to value, built for lean teams without dedicated IT support
Back to top