Rolling forecasts for real estate: how to forecast when rent and debt drive everything

Annual budgets break down fast in real estate. Not because the numbers are "wrong," but because the timing changes:

• a void drags on for two extra months
• a refinance slips by a quarter
• rates move 100-200 bps
• a refurb programme accelerates (or stalls)
• arrears spike briefly, then unwind

In most portfolios, a handful of lines determine whether you are comfortable or stressed:

1. Rent (and occupancy)
2. Debt service (and rates)
3. Capex timing
4. Cash runway + covenants

That is why rolling forecasts work better than static budgets: they are built to absorb change-without rebuilding the model every time.

This post explains a practical way to run a rolling forecast when rent and debt drive everything, and what to focus on so your forecast stays usable month after month.

---

What a rolling forecast is in real estate terms (not FP&A theory)

A rolling forecast is a forecast horizon that "rolls forward" as time passes.

Instead of locking a budget once per year, you update the forecast at a regular cadence (usually monthly), so you always have visibility for the next 12-24 months (or whatever horizon matters for your lenders and liquidity).

For real estate, a rolling forecast is less about "perfect P&L prediction" and more about answering these questions continuously:

• How much cash will we have, and when?
• Will we breach covenants (DSCR/ICR/LTV), and in which SPVs?
• What is our funding requirement if we accelerate capex or occupancy drops?
• How sensitive is the portfolio to rates and voids?

If your forecast can answer those four, it is doing its job.

---

Why rent + debt dominate (and what that means for your forecast design)

In a typical property SPV, rent is the primary inflow and debt service is the primary outflow. Most other lines matter, but they rarely swing the outcome as violently as:

• a change in occupied units or passing rent, or
• a change in interest rate / refinancing terms / amortisation, or
• capex timing that changes cash at the worst possible time

So the smartest rolling forecasts are driver-based and built around two engines:

1. a rent engine (leases -> occupancy -> rent -> collections)
2. a debt engine (principal -> rate -> interest -> covenants -> cash lock-ups)

Everything else sits around those engines.

---

The real estate rolling forecast playbook

Step 1: Pick the horizon that matches your risk window

Most teams land on:

• 13 weeks (cash only, tactical)
• 12 months (operational planning)
• 18-24 months (refinancing + covenant visibility)

A common setup is a dual horizon:

• a high-confidence near-term view (next 3 months), and
• a model-driven medium-term view (months 4-18/24)

The right answer depends on how often you refinance, how rate-sensitive you are, and how quickly you can adjust capex.

---

Step 2: Forecast cash first, then reconcile profit

This is a mindset shift that saves time.

If you start with a P&L-only forecast, you will end up rebuilding it to answer cash questions anyway. Instead, build (or run) a forecast that produces:

• NOI (for performance)
• Debt service (for survival)
• Capex timing (for liquidity)
• Cash balance (for decision-making)

Then layer P&L nuance (accruals, prepayments, timing differences) as needed.

---

The rent engine: forecasting income the way properties actually behave

1) Start with the rent roll (not a growth assumption)

A real estate forecast becomes credible when it reflects contracted reality.

Build your rent forecast from:

• current passing rent by unit/lease
• lease start/end dates
• escalation clauses (annual uplift, indexation, step-ups)
• known renewals (if already agreed)
• known void periods (refurb downtime, lease expiries, major works)

Even if you do not model each unit, you can model cohorts:

• in-place leases
• expiring within 3 months
• expiring within 12 months
• vacant units
• refurbished vs unrefurbished stock

This lets you forecast occupancy and rent changes as events rather than smooth averages.

---

2) Forecast occupancy as a timeline, not a single %

Occupancy changes over time, and it usually changes in steps:

• a lease ends
• a unit goes offline for works
• a tenant moves in after a lag

So rather than "92% flat," build a simple monthly series:

• opening occupied units
• move-outs / expiries
• expected void period
• move-ins
• closing occupied units

Then derive:

• physical occupancy
• economic occupancy (if concessions/arrears matter)

Tip: you can keep this lightweight. The goal is not micro-precision; it is capturing the timing of the big steps.

---

3) Separate billing from collections (especially in stress cases)

Many forecasts fail because they assume:

billed rent = cash received

In reality, collections lag, and arrears can spike.

A practical approach is to add a collections assumption:

• 98-99% in base case
• lower during known tenant stress periods
• a catch-up profile if arrears are expected to recover

This single layer often explains why a portfolio "looked fine on NOI" but ran tight on cash.

---

The debt engine: forecasting the line that can flip your cash outcome

1) Model debt by tranche and by type

At minimum, each facility needs:

• opening balance
• amortisation / repayment schedule
• maturity / refinance date
• fixed vs floating
• margin and fees
• DSCR/ICR/LTV definitions (or at least proxy metrics)
• reserve requirements (DSRA, cash traps, minimum liquidity)

Even a "simple" debt engine should let you answer:

• What is interest next month?
• What is interest 12 months from now if rates change?
• What happens to cash if we refinance late?

---

2) Treat interest rates as scenarios, not a single number

Rates move. Your forecast should assume that they will.

A practical way to do this without over-engineering:

• Base rate path (your best estimate; can be flat, or a curve you update)
• Rate shock scenarios (e.g., +100 bps, +200 bps, -50 bps)
• Optional: hedge/cap logic if you have it

Then the debt engine can compute:

• all-in rate
• interest cost
• coverage metrics
• cash impact (including any reserve top-ups)

---

3) Put refinancing risk on the dashboard

Refinancing is where rolling forecasts earn their keep.

Add a simple refinance view:

• maturity date
• assumed refinance date (with slip scenario)
• refinance fees
• new margin assumption
• any required paydown

Then surface:

• minimum cash balance under "refi slip"
• covenant headroom under "higher margin"
• peak funding requirement if refinance delays

---

Opex and capex: keep it simple, but tie it to the plan

Opex: forecast with a "fixed + variable + known events" structure

Rather than line-by-line perfection, use:

• fixed costs (insurance, certain contracts, PM fees if fixed)
• variable costs (utilities, turnover costs, letting fees)
• known events (one-off compliance spend, planned contractor works)

For many portfolios, this is accurate enough-because rent and debt dominate the swing.

---

Capex: forecast timing explicitly (timing matters more than totals)

Capex is often the difference between "fine" and "tight" in the next 6-12 months.

For a rolling forecast:

• model capex monthly (not annual)
• connect capex to refurb/works programme cadence
• include downtime impact if it affects lettings
• include contingency and slippage scenarios

If you only do one thing: get capex timing right.

---

The monthly rolling forecast process that keeps teams sane

A rolling forecast only works if it is operationally repeatable. Here is a practical monthly cadence:

Month-end (Actuals close)

1. Lock the period and confirm reconciliations
2. Refresh actuals into the forecast model/reporting layer
3. Investigate variances: what changed vs last forecast?

Forecast update (1-3 days later)

4. Update rent roll changes (move-ins/outs, renewals, reversion assumptions)
5. Update debt assumptions (rates, refinance timing, covenant thresholds)
6. Rephase capex based on the real programme
7. Run scenarios (rates, occupancy, capex acceleration)

Output pack (the bit stakeholders actually want)

8. Publish:

• base forecast
• downside case (rates/occupancy)
• cash runway chart + covenant headroom
• commentary: what moved and why

Key rule: every month you should be able to say, in plain English:

"Cash is down because capex pulled forward and occupancy recovered slower than planned-rates added X on top."

---

Common mistakes (and how to avoid them)

Mistake 1: Forecasting NOI but not cash

Fix: always produce a cash roll-forward and minimum cash balance.

Mistake 2: One occupancy number for everything

Fix: treat lease expiries, downtime, and void periods as timeline events.

Mistake 3: Ignoring refinancing and reserve mechanics

Fix: include maturity timing, fees, and reserve top-ups as explicit cash items.

Mistake 4: Manually rebuilding spreadsheets every month

Fix: separate the "engine" from the "assumptions," and update via a structured input layer (rent roll, debt assumptions, capex plan).

Mistake 5: No version control or assumption register

Fix: keep a simple "what changed this month" log (rates, occupancy, capex phasing, refinance assumptions).

---

What "good" looks like: the outputs that earn trust

A real estate rolling forecast is doing its job when you can produce:

• Portfolio + SPV cash runway (min cash balance, when and where)
• Debt service + coverage (DSCR/ICR proxy) with breach months highlighted
• Occupancy and rent bridge (what drove revenue changes)
• Capex phasing view (plan vs latest forecast)
• Scenario impact summary (rates/occupancy/capex timing -> cash impact)

If your stakeholders can drill into "why" without a spreadsheet forensic exercise, you have built something scalable.

---

How we help (and why this matters for multi-SPV portfolios)

Rolling forecasts get dramatically easier once the portfolio is consistently structured:

• one-stop visibility across multiple Xero or QuickBooks SPVs
• standardised charts of accounts and mappings so entities roll up cleanly
• FP&A for budgeting, forecasting, portfolio/SPV reporting, and cash planning
• scenario planning (rates, occupancy, refurb programmes) with cash impact
• real estate-specific metrics (occupancy, NOI, yields, gearing)
• automated reporting and narrative commentary for boards/investors