Skip to content

Bridging - Fundamentals

What It Is (General)

Bridging in treasury refers to the reconciliation process that explains the variance between direct cash flow forecasts (transaction-based, showing actual cash movements) and indirect cash flow forecasts (derived from accrual-based P&L budgets and balance sheet changes). The goal is to identify and categorize the drivers of any differences, typically working capital movements, timing differences, or forecast errors.

What It Means for Our ICP

How Treasury Teams Think About It

Treasury teams view bridging as the essential "translation layer" between their operational cash reality and the financial planning world. They need to answer questions like: "We budgeted X EBITDA, but our cash position is Y—why?"

The core mental model involves: 1. Starting with indirect forecast (budget EBITDA) 2. Adjusting for non-cash items 3. Adding/subtracting working capital movements (AR, AP changes) 4. Layering in interest, capex, treasury items 5. Arriving at a number that should match direct cash flows

Euroports' Process (Source: 2025-10-27) 1. Classify direct cash flows from bank transactions 2. Assign GL accounts based on budget structure 3. Aggregate in Excel 4. Compare to budget EBITDA 5. Derive working capital movement as the "plug"

Typical Processes & Timing

  • Frequency: Bi-weekly bridging analysis is common for active treasury teams
  • Timing: Often aligned with management reporting cycles
  • Challenges: ERP bookings only updated monthly, making within-month bridging difficult

Tools They Use Today

  • Cash Analytics - TMS for transaction data, but lacks drill-down and variance analysis (Mentioned by: Euroports)
  • Excel - Primary tool for actual bridging calculations and variance analysis (Mentioned by: Euroports)
  • ERP systems - Source of GL/budget data, but often updated too infrequently

How They Talk About It

  • "Bridging direct vs indirect" - the core activity
  • "Variance drivers" - what they're trying to identify
  • "Working capital movement" - often the largest unexplained variance bucket
  • "Pollution of categories" - when classification errors corrupt reporting
  • "Deviation from budget" - the gap they need to explain
  • "Reporting hub" - desired centralized place for all variance analysis (Personio)
  • "GFC" (Rolling Forecast) - monthly updated forecast more accurate than static budget (Personio)
  • "Operating vs Non-Operating" - FP&A split not replicable from bank statements (Personio)
  • "Cash Engine" - FP&A revenue model with assumptions and seasonality (Personio)
  • "T+1 analysis" - first-day-of-month cash burn reporting (Personio)
  • "Last Estimate" / "Previous Estimate" - versioned forecast labels for board comparison (Euroports)
  • "MPL vs Non-MPL" - Matthias' split between freight forwarding (summarized) and terminals (decomposed) (Euroports)
  • "Structural / stable / unfavorable variance" - variance classification Matthias wants bucketed automatically (Euroports)

"We want to know the deviation of the short term forecast versus the budget... I want to understand the drivers." - Matthias, Euroports

"What I'm aiming to achieve is being able to use the system as like, a true reporting hub." - Tom Thorn, Personio

"There's the actual versus direct forecast, there's the actual versus indirect forecast, and then there's the forecast versus forecast." - Tom Thorn, Personio

Personio's Process (Source: 2025-12-04) 1. Receive FP&A monthly budget/GFC numbers 2. Manually split into weekly transactional forecasts 3. Compare three variance points: actuals vs direct, actuals vs FP&A, forecast vs forecast 4. Flag significant variances and investigate 5. Feed back explanations to FP&A team

ON's Bridging Need (Source: 2026-03-04) Lucia describes controlling's 5-year P&L plan as "just the consequence of a lot of other planning" — hard to assess realism. She uses Palm's direct cash data as an independent sense check. The bridge is not just about explaining variance, but about validating whether indirect assumptions translate to realistic direct cash flows. BigQuery/ERP integration is prerequisite for this to work at full power.

Euroports' Board Presentation Format (Source: 2026-04-15) Always by quarter. Columns shown on every row: - Actual (will come from Palm) - Last Estimate (latest forecast) - Previous Estimate (prior forecast — requires version locking) - Prior Year - Budget

Structure: EBITDA variance → EBITDA adjustments → taxes & management fees → working capital movement (split MPL vs Non-MPL) → capex → finance items. Board mostly looks at the delta vs budget (Column E), not the absolutes. Matthias fills a free-text commentary box per row explaining the variances (collections delays, factoring penalties, overdue invoices, etc.).

MPL vs Non-MPL split: MPL = freight forwarding (treated as one line, no drill-down); Non-MPL = terminals (fully decomposed with variance per line and entity breakdown). Two different ERPs feed each (Oracle for terminals, InVision for freight forwarding).

EBITDA mechanics: EBITDA + EBITDA adjustments come from budget directly — no direct equivalent. Revenue (line 6) minus certain costs (line 10) = EBITDA. If local entity changes row 6 or 10, EBITDA auto-updates. Working capital rows split: trade receivables, trade payables, intercompany (should net to zero consolidated), and "other" (accruals — the hardest bucket to explain).

Euroports' Forecast Bias Management (Source: 2026-04-15) "Depending on who does a forecast, their personal bias influences that. If somebody from AP is doing it, obviously they overestimate payments because they want to make those payments." Group treasury compensates by manually adding target overlays — e.g., "Spain, add 1M collections in March" — to force local forecasts toward realistic numbers. Plan is eventually to surface forecast accuracy KPIs per entity as a CFO-visible scoreboard (not to shame, but as "carrot and stick" to drive improvement). Complements bridging because accurate local forecasts reduce the variance that has to be explained after the fact.

FP&A-Treasury Relationship

Levi's Perspective (Source: 2025-12-11) - Cadence difference: FP&A refreshes annually/quarterly, Treasury refreshes weekly - Permanent vs timing differences: Track which variances are timing (will resolve) vs permanent (need explanation) - Who knows first: Treasury often knows real numbers before FP&A (e.g., acquisition costs, disruptions) - Working capital hedge: FP&A maintains a "cushion" or range; Treasury feeds information to calibrate it - Acquisitions example: "Once negotiations are on the table, I have a better number. They need to follow my number." - Information = success: "We usually are very successful in achieving the goal because of that information."

ON's Two-Tier Validation Framework (Source: 2026-04-29) Lucia frames forecast confidence as resting on two legs:

  1. Variance analysis — Palm's forecast vs Palm's past actuals. Looks backward. Already supported by variance-analysis features. "Was it hitting or coming close to the actuals."
  2. Plausibility check — Palm's forecast vs controlling's long-term plan. Looks forward. Today done manually in Excel/Anaplan as a gut-feel exercise. "Does my forecast more or less match what controlling is saying we're going to hit?"

The bridge is fundamentally an automated plausibility check. Without leg #2, treasury hesitates to commit to acting on the forecast — controlling's plan is the binding language of the organization for >13 weeks, even when treasury individually disagrees with the assumptions.

ON's Time-Horizon Split (Source: 2026-04-29) Lucia is explicit that bridging the two forecasts works differently at different horizons:

  • ≤13 weeks: Palm = source of truth. Operational, treasury-driven. Long-term plan should NOT bias the short-term forecast. "You come up with a better forecast by ignoring the long term in the short term."
  • >13 weeks (especially >6–7 months): FP&A long-term plan = source of truth. Palm should incorporate it as a high-quality assumption. "The longer term we go, the higher the weight of the forecast from controlling takes."

Implication for product: the bridge view should not present the comparison as "your plan is wrong" — it should respect that controlling's plan binds the long-horizon, while Palm's forecast binds the short-horizon, and the conversation is about consistency in direction rather than convergence in numbers.

ON's Bridge Translation Mechanics (Source: 2026-04-29) Anaplan bridge logic = calculations + maintained tables. Calculations are self-sustaining; tables are high-touch (require constant care + re-justification each cycle). Examples Lucia showed live: - AR aging splits per company — e.g., "70% paid within one month for entity X" (table) - VAT timing assumptions — when VAT becomes payable - Inventory→COGS→cash chain — P&L recognizes COGS at point of sale, but cash is paid for inventory bought earlier; Anaplan models this with an inventory-movement detour and payment-time tables

Lucia's hypothesis: most of these can be replaced by statistical inference from historical data. "I think this is where statistics do a much better job than we can do." The inventory→COGS detour is her favorite example — she'd rather Palm correlate warehouse cash payments to sales growth directly.

ON's "Low-Touch" vs "No-Touch" Distinction (Source: 2026-04-29) Lucia carefully separates two desirable states: - No touch — model runs autonomously; user just confirms it looks right ("yep, sounds about right") - Low touch — user can override an assumption when they have specific knowledge ("nah, let me type in five instead of two")

She wants day-to-day operation to be no-touch (alerts only when something needs attention). But the override path must remain for special-knowledge cases (e.g., a celebrity collab driving a quarter, a one-off market event). Both states require explainability — without seeing the model's reasoning, treasury won't trust the no-touch state and won't know when to override.

ON's Persona Breadth for the Bridge (Source: 2026-04-29) The bridge is explicitly not a treasury-only tool. Lucia identifies three users: 1. Treasury — primary user; uses bridge as the plausibility check 2. Controlling — joint user; gets a cash-bridge angle to add to their existing P&L actual-vs-forecast challenges to regional teams (e.g., APEC region "always off because we grow faster than expected") 3. CFO — tertiary; "wants to understand how my P&L transforms into [cash]" and look at the insights directly without going through treasury

"It's a conversation that's probably half treasury, half controlling. But super happy to go into this in-between area."

Controlling's Plan as Binding Source of Truth (Source: 2026-04-29) Lucia is explicit that even when individual contributors (including her) disagree with assumptions in the consolidated long-term plan, the plan is binding once consolidated. "I can look at the sales [forecast] and say, hey, this is completely unrealistic. But hey, sales team said that. So. Okay." This shapes how Palm should present the bridge — not "your plan is wrong" but "here's the gap between your plan and the cash reality, here are possible drivers." Palm needs to respect controlling as the canonical owner of long-term assumptions, even while Palm provides the better mechanical translation to cash.

Sources: view all