Budgeting for refurb and capex: how to avoid surprises in month 6

Why capex plans feel "on track" in months 1-3... then blow up in month 6-and how to build a budget that stays real.

Refurb and capex programmes rarely fail because the total number was wrong on day one. They fail because timing, scope, and cash reality drift-quietly-until month 6, when the surprises show up:

• invoices land later than expected (and in bigger chunks)
• variations and "unforeseen" items stack up
• voids last longer than your letting assumptions
• project pace slips by just enough to hit the wrong season
• cash gets tight right when debt service is rising

The good news: you do not need a perfect crystal ball. You need a budgeting approach that treats refurb/capex like what it is:

A delivery programme with cash timing, operational dependencies, and risk-not just a line item.

This post lays out a practical method to budget refurb and capex so you can see pressure building early, not when you are already in month 6.

---

Why "month 6" is the danger zone

Month 6 surprises usually come from three lags that your initial budget did not model properly:

1) The commitment-to-cash lag

You approve a contractor quote in month 1, but cash might not leave your account until months 4-6 because of:

• mobilisation schedules
• staged valuations
• retention
• invoice batching
• procurement lead times

Your spreadsheet says "capex this quarter." Reality says "capex when it hurts most."

2) The works-to-income lag

Even when capex is delivered, income recovery is delayed:

• unit downtime extends
• leasing demand is seasonal
• rent premiums take time to realise
• renewals happen later than the original plan

So you get the cost now, but the uplift later.

3) The scope-to-variance lag

Variations rarely hit immediately. They appear after:

• strip-out reveals issues
• compliance and building control feedback arrives
• "while we are here" upgrades get added
• specification creep happens one decision at a time

By month 6, the reforecast becomes unavoidable-and painful.

---

The principle: budget capex like a pipeline, not a pot

If you want fewer surprises, stop thinking:

"We have -X of capex this year."

Start thinking:

"We have a pipeline of projects with committed costs, expected cash timing, delivery risk, and income impact."

That shift changes how you build the budget and how you monitor it.

---

Step 1: Split capex into clear buckets (so reporting stays consistent)

Most "capex surprises" are actually classification surprises-spend that moves between opex and capex, or between refurb and lifecycle, depending on who coded it.

A practical capex budget uses buckets that match how decisions are made:

Refurb / value-add (business plan capex)

• unit upgrades, repositioning, rent premium works
• typically linked to downtime and leasing assumptions

Lifecycle / maintenance capex

• roofs, boilers, lifts, plant replacement
• less upside, but non-negotiable

Compliance / safety capex

• fire safety, EICRs, cladding-related works, accessibility
• often schedule-driven and risk-driven

ESG / efficiency capex

• insulation, heat pumps, LED, controls
• may be linked to grants, EPC targets, lender requirements

Why this matters: each bucket behaves differently in cash timing and forecasting confidence. Treat them the same and your month 6 reforecast becomes a mess.

---

Step 2: Build the refurb plan from cadence + downtime (not just "-/unit")

A refurb budget that avoids surprises has three drivers:

1. Cadence (units per month / phases per quarter)
2. Downtime (how long units are offline)
3. Rent premium timing (when uplift actually hits cash)

A simple refurb pipeline structure

Even in a spreadsheet, you can model:

• Units started (per month)
• Units offline (under works)
• Units completed (returning to service)
• Cumulative refurbished units
• Capex cash out (by stage)

This turns refurb into a forecastable production line rather than a single annual number.

Pro tip: separate "start of works" from "cash paid." Many programmes look fine until invoice timing catches up.

---

Step 3: Budget cash using three layers: Planned, Committed, Forecast

To avoid month 6 shocks, your capex reporting needs to show where you are in the spending lifecycle, not just actuals.

Layer A: Planned (the budget)

What you intended to do this month/quarter.

Layer B: Committed (the reality starting to form)

Costs you have agreed to pay but have not paid yet:

• signed contracts
• approved purchase orders
• accepted quotes
• variations approved in principle

Committed spend is the early warning system.

Layer C: Forecast (latest best estimate)

Planned adjusted by:

• slippage
• scope changes
• procurement updates
• delivery constraints

If you only track "Actual vs Budget," you will always find out late.
If you track "Committed vs Budget," you find out early.

---

Step 4: Add the two contingencies people forget

Most teams include some contingency. They often miss the two that actually bite.

1) Variations contingency (scope risk)

Typical drivers:

• latent defects
• spec upgrades
• compliance requirements
• access constraints

Practical approach:

• set a bucket-level contingency (refurb - lifecycle - compliance)
• release it only with documented reasons
• log every drawdown (so contingency does not quietly become "extra budget")

2) Timing contingency (cashflow risk)

Even if total spend stays the same, timing can kill you.

Add a "timing buffer" by modelling:

• a slower pace scenario (e.g., refurb cadence drops 30%)
• an invoice bunching scenario (e.g., 2 months of capex paid in one)

Then view the impact on:

• minimum cash balance
• covenant headroom
• funding requirement

This is how you avoid discovering in month 6 that you were "right on total, wrong on timing."

---

Step 5: Connect capex to the cash runway and debt covenants

Capex budgets often live in project land while debt lives in finance land. Month 6 is where they collide.

Every capex plan should feed a simple set of finance outputs:

• Minimum cash balance (by SPV and portfolio)
• Peak funding requirement (if cash goes negative)
• Debt service coverage / interest coverage (even if it is a proxy)
• Reserve movements (if you operate DSRA or cash traps)

If your reporting does not show when cash gets tight, the budget is not decision-grade.

---

Step 6: Run a monthly capex reforecast cadence that is lightweight

You do not need bureaucracy. You need rhythm.

A practical month-end cadence:

1. Close actuals (and lock the period)

2. Update committed capex (contracts/quotes/variations)

3. Rephase the next 3 months based on the live programme

4. Reforecast months 4-12 using updated cadence and procurement timing

5. Publish:

   • "Budget vs Actual vs Committed vs Forecast"
   • cash runway + covenant headroom
   • a short commentary: what changed, why, what is at risk

This is how you prevent month 6 surprises: you surface drift in month 2-3.

---

Common causes of month 6 surprises (and how to neutralise them)

Surprise: "Capex was coded to opex (or vice versa)"

Fix: enforce a portfolio-level classification policy and keep entities aligned (especially across SPVs).

Surprise: "Downtime was underestimated"

Fix: model downtime explicitly and maintain a "void reason" log (works vs leasing vs tenant delay).

Surprise: "Procurement lead times shifted the whole schedule"

Fix: track long-lead items separately and add timing contingency.

Surprise: "Variation approvals were not visible until invoices arrived"

Fix: treat approved variations as committed immediately-even if not invoiced.

Surprise: "The uplift did not hit when we expected"

Fix: separate:

• works complete date
• back-to-market date
• lease signed date
• first cash received date

Those four are rarely the same month.

---

What "good" looks like: the capex dashboard that stays trusted

A capex budget becomes operational when you can answer, at any time:

• What have we spent?
• What have we committed?
• What will we spend next, and when?
• Which assets/SPVs are drifting, and why?
• What does the latest plan do to cash and covenants?

If you cannot drill into those questions, month 6 will always be a surprise.

---

How we help (and why this is hard across SPVs)

Capex budgeting gets exponentially harder when it is spread across multiple SPVs-different charts of accounts, different coding habits, different project trackers, and different month-end quality.

That is why our platform is designed around multi-entity consolidation for Xero or QuickBooks SPVs, standardised charts of accounts and mappings, and portfolio dashboards that drill from portfolio -> SPV -> account, with FP&A and scenario planning built in.

When your data model is consistent, capex stops being a month-end detective story-and becomes a forecastable programme tied directly to cash.