How to Build an Investor-Ready Financial Model in 90 Days

Owner-managed groups in the Gulf often come to me with the same request: 'We need a financial model for the investors.' What they usually have is a worksheet someone built three years ago, with hard-coded numbers, a single optimistic scenario, and no balance sheet. What they need is a coherent quantitative argument about the business — defensible at every line, flexible enough to absorb diligence questions, and consistent with the audited IFRS financials a sophisticated investor will reconcile to within the first hour.

I have built investor-ready models for owner-managed businesses in trading, contracting, industrial services, and manufacturing across Saudi Arabia and Bahrain. The work always sequences the same way, and 90 days is the right amount of time. Less than that and the model is fragile. More than that and the business has moved on and you are modelling history. Below is the sprint-by-sprint structure I use.

Sprint 1 (Days 1-15): Foundations and historical truth

Nothing forward-looking gets built until the historical foundation is solid. The first sprint is unglamorous and absolutely critical.

1. 01Pull the last three to five years of audited IFRS financial statements at the entity level for every legal entity in the group.
2. 02Build a consolidation file that ties to the audited consolidated financials, with intercompany eliminations explicit and traceable.
3. 03Rebuild monthly management accounts for the trailing 24 months from source — not from the existing management report, from the underlying ledger and bank data.
4. 04Construct the EBITDA normalisation bridge from reported to underlying, with every adjustment categorised, quantified, and supported by a working note.
5. 05Build a working-capital decomposition: debtor days by customer category, inventory days by stock category, creditor days by supplier category.

By the end of Sprint 1 you should have a single workbook — I call it the Historical Pack — that an investor's analyst could open and reconcile to your audited statements within an hour. If they can't, every forward-looking number you build on top of it is on sand.

Sprint 2 (Days 16-30): Architecture and revenue build

Now the model itself. The architecture choice you make in this sprint determines whether the model can survive the diligence process or collapses under the first round of investor questions.

Architecture principles

I build every investor-ready model with the same skeleton: separate tabs for assumptions, calculations, and outputs; one row per line item with no merged cells; consistent time periods across every sheet; and a strict rule that no output number is ever hard-coded. Every cell in every output traces back through formulas to a single source assumption.

Revenue build

Revenue is built bottom-up, never top-down. The structure depends on the business model:

• Contracting and project businesses: backlog plus pipeline, with each contract or expected contract modelled by phase, with associated billing and collection profile.
• Trading and distribution: customer-by-customer for top 20 customers, category-level for the long tail, with explicit assumptions on price and volume separately.
• Industrial services: contract-by-contract for recurring agreements, with utilisation-based modelling for any capacity-driven revenue.
• Manufacturing: SKU-level or product-family-level volume forecast multiplied by price, with capacity constraints made explicit.

In every case, the revenue tab must reconcile to the historical actuals for the last 24 months before any forward-looking numbers are added. If your build can't reproduce history within a few percent, the build logic is wrong, not history.

Sprint 3 (Days 31-45): Cost structure and three-statement integration

Cost structure is where most owner-built models fall apart. The instinct is to model costs as a percentage of revenue. That works for a single year of forecast in a stable business. It does not work for a model that needs to defend itself across multiple scenarios over five years.

The right approach is to split costs into four categories and model each on its own driver:

1. 01Direct variable costs — modelled per unit of revenue activity, with explicit gross margin assumptions by product or service line.
2. 02Direct fixed costs — modelled by site, asset, or contract, with step changes when capacity thresholds are crossed.
3. 03Indirect costs (overhead) — modelled by department or function, built up from headcount and unit costs, not as a percentage of anything.
4. 04Capex and depreciation — modelled from the asset register, with maintenance vs. growth split explicit, and depreciation calculated by asset class with appropriate IFRS-compliant useful lives.

Once the P&L is built this way, the next step in this sprint is integrating the balance sheet and cash flow. A model without a working balance sheet is not an investor-ready model — it is a forecast P&L with cash flow attached. Every investor will run the integrity checks: does the balance sheet balance every period, does the cash flow tie to the change in cash, does retained earnings roll correctly. If any of those fail, you lose credibility for the rest of the conversation.

Sprint 4 (Days 46-60): Working capital, debt, and tax

By Sprint 4 the bones of the model are in place. This sprint is about making it behave like the actual business.

Working capital

Working capital is modelled as days outstanding by category, applied to the relevant P&L driver. Debtor days against revenue (with adjustments for VAT in Saudi and other applicable indirect taxes), inventory days against cost of sales, creditor days against operating costs. Crucially, the working-capital schedule must reflect the actual seasonality of the business — most GCC trading businesses have material Ramadan and summer effects that a flat-line model will miss completely.

Debt schedule

Every facility from the facility map gets its own row in a debt schedule. Drawdowns, repayments, interest or profit charges, and ending balance, period by period. Where facilities are revolving, the model needs explicit utilisation logic linked to the working-capital cycle. Where there are covenants, those covenants are computed every period in the model, with headroom shown explicitly. An investor will model your covenants whether you do or not — better that they see your version first.

Tax and zakat

Saudi groups face zakat at 2.5% of the zakat base for Saudi/GCC ownership and corporate tax at 20% on non-GCC ownership. Bahraini entities are largely tax-free for now, but VAT and other levies still apply. The model needs to reflect the right structure for the right entities, and any group-level tax planning needs to be visible, not buried.

Sprint 5 (Days 61-75): Scenarios and sensitivity

A model with one scenario is not an investor-ready model. It is a slide. By the end of Sprint 5 the model should run a minimum of three named scenarios — Base, Downside, and Upside — with each scenario driven by changes in a small set of named input variables, not by manually overwriting numbers across the model.

I typically build a scenario manager tab where the user picks a scenario from a drop-down, and the model recalculates every output cleanly. The variables that matter most differ by industry but usually include some combination of: revenue growth rate, gross margin, working-capital days, capex intensity, and discount rate.

Sensitivity analysis is separate from scenarios. A sensitivity table shows how a single output (usually equity value or DSCR) responds to changes in two key inputs at a time. Investors expect to see sensitivity tables on revenue growth × gross margin, and on exit multiple × terminal growth rate. Build them, present them, and be ready to explain why the curves bend the way they do.

Sprint 6 (Days 76-90): Output pack and stress testing

The final sprint is about turning the model from a working file into a presentation-grade artefact.

1. 01Build a one-page output dashboard with the key financial metrics by year — revenue, gross margin, EBITDA, EBITDA margin, capex, free cash flow, net debt, leverage ratios.
2. 02Build a valuation tab — DCF with explicit terminal value, plus comparable trading multiples and precedent transaction multiples relevant to the sector and the Gulf.
3. 03Build a returns tab if the conversation is with private equity — IRR, money multiple, and the bridge from entry equity to exit equity decomposed into operating improvement, multiple expansion, and leverage.
4. 04Stress test every formula in the model. Break things on purpose, see what cascades, fix the cascades.
5. 05Have someone other than the model builder rebuild a single key calculation from the inputs and verify it matches. If it doesn't, the model isn't ready.

The final deliverable is the model itself, plus a short user guide explaining where to find what, plus the historical pack from Sprint 1, plus the EBITDA normalisation bridge as a standalone document. Together, these four artefacts are what gets sent to the investor or the buyer when the conversation gets serious.

"An investor-ready model is not the one with the most tabs. It is the one where every number can be defended, every scenario can be re-run, and every integrity check passes on the first try."

What 90 days actually looks like in practice

I want to be honest about the resourcing. A model of this calibre, for an owner-managed GCC group with two to four entities, typically takes one senior modeller working three to four days a week for the full 90 days, with input from the owner and finance team probably equating to one full day a week of their time. Trying to compress it to 30 days, or trying to do it with a junior analyst alone, almost always produces a model that fails diligence.

It is also work that pays for itself many times over. In the last two years I have built models that supported a SAR 140 million growth equity raise for a Saudi industrial services group, a partial sale of a Bahraini distribution business, and several refinancing packages where the model was the document that got the credit committee comfortable. In each case, the model itself was not the deal — but the deal would not have happened without it.

When to start

If you are within six months of needing to show a model to an investor, an acquirer, or a lender for a significant facility, you should be starting the 90-day build now. The model is not the last thing you do before you go to market. It is the first thing, and everything else — the information memorandum, the management presentation, the diligence responses — flows out of it.

At Aontas Advisory we build investor-ready models for owner-managed businesses across the Gulf as part of our fractional CFO engagements, often as the spine of a broader capital raise or transaction process. If you are looking at a fundraise, a sale, or a major facility conversation in the next 6 to 12 months, our Fractional Executive Services page sets out how we work with owners through this kind of build.