What's inside

Eleven things arrive. Here is what each one produces.

No locked cells, no hidden sheets, no macros, no password. Every formula is written out in English beside the cell it drives, because you cannot defend a number in a client meeting if you cannot open it.

The workbooks

Six files that do the work.

XLSX · 13 tabs

Master planner

Brief intake, assumptions with sources, market scoring, platform scoring, budget allocation, forecast, scenario planner, commercial economics, monthly phasing, campaign structure, client export, QA notes.

Produces: a plan, a forecast, a conservative case, and a written reason for every market and platform decision including the exclusions.

Fragmentation flag per lineSpend to contribution marginStages you can switch off
XLSX · 4 calculators

Commercial calculators

Affordable CPL, breakeven ROAS, fragmentation, funnel sensitivity. They run entirely on the client's own economics. No industry data enters the calculation at any point.

Produces: the cost per lead this business can actually support, and the answer to which single rate, if it moved a tenth, would change the outcome most.

Sensitivity tableCells you could affordTrue ROAS after over-attribution
XLSX · 47 checks

Activation QA

Account and commercial, campaign setup, creative, measurement, launch governance. Plus a launch tracker with owners, deadlines, blockers and severity.

Produces: a go or no-go verdict. It refuses to clear you while any check is marked pass with no evidence recorded, which is deliberate and occasionally annoying.

Filter by platformSeverity levelsEvidence columnTasks with no owner counter
XLSX · 11 templates

Reporting system

Weekly, monthly, executive summary, campaign wrap, test and learn, decision register, action tracker, creative log, market comparison, platform comparison, and a commentary library.

Produces: a report structured around five questions in order, ending in a decision. Plus twelve situations with the version written at 6pm and the version that prevents a meeting.

Does it matter columnWhat we assumed at the time12 commentary situations
XLSX · 8 tabs

Pacing and invoice reconciliation

Approved plan, PO register, platform actuals, accruals, invoices, reconciliation and a dashboard, with a documented source-of-truth hierarchy.

Produces: seven automatic flags, in words you can paste into an email. Spend over PO, invoice over actual, actual with no PO, and the one that surprises everybody: spent, not invoiced, not accrued.

Outstanding exposurePO expiry trackingFX and fee handlingVariance by market
XLSX · 4 tabs

Source register

Every time-sensitive claim with its source, the source's own effective date, the date it was last checked, a confidence level and a revalidation flag. Plus planning ranges and a verification log.

Produces: provenance. Every claim carries a source, a date and a confidence label, so you can check the work rather than take it on trust.

Two dates per claimConfidence labelsMethod, where no number is defensible
The reasoning

The part that makes the workbooks worth opening.

Decision logic, not education. Every section answers the same eight questions: what decision, what information, what rule, what goes wrong, what tool, what output, what to tell the client, what risk to record.

Brief diagnosis

Thirteen flags. For each: why it matters, the exact question to ask, the decision that depends on it, the assumption you can hold temporarily, and the risk to record.

Four market playbooks

UAE, Saudi Arabia, Kuwait and Qatar. Strategic role, language, sales coverage, lead routing, seasonality, measurement risk, and when to deprioritise. Questions, not country facts.

Eight platform cards

Meta, TikTok, Snapchat, Google Search, YouTube, programmatic, connected TV, LinkedIn. What each is genuinely useful for and what it is routinely forced to do badly.

Funnel and KPI logic

Business outcome to customer behaviour to optimisation event to KPI. Plus the conversion value ladder, and why full-funnel is a budget size rather than a best practice.

Measurement planning

A ten-field plan, a nineteen-item readiness checklist, and the WhatsApp handoff problem stated plainly rather than discovered in month two.

Optimisation diagnosis

Five problem classes with symptoms, checks in order, likely causes, what not to change yet, evidence required, and client-safe wording for each.

Four vertical cards

Automotive, real estate, luxury, beauty. What changes for that vertical and nothing else. They modify the system rather than replacing it.

AI prompt pack

Sixteen prompts built so the model argues against your plan instead of agreeing with it, and writes MISSING rather than inventing a number.

Ten Louli Reality Checks

Each one a 60-second check against a live plan, with a decision rule and a tool. The uncomfortable one is usually the one worth acting on.

The proof

Four fully worked GCC campaigns.

Brief in, reconciled invoice out, every major decision annotated with the arithmetic shown. Fictional brands, altered numbers, and no industry benchmark used at any point.

AED 75,000 · UAE · lead generation

The client's AED 60 target was real and measuring the wrong thing. The business could afford AED 102. The cheapest fix of the whole campaign was a phone number field that had been silently rejecting valid submissions.

Read this one in full, free

USD 100,000 · KSA and UAE · launch

No account history, so every rate in the forecast is borrowed. The deliverable becomes the range and the re-forecast date rather than the point estimate, and saying that out loud is the job.

USD 250,000 · GCC · full-funnel

Even at a quarter of a million, three stages across four markets does not clear the floors. Two does. Includes the brand lift decision, and the finding that sits in the footnote rather than the table.

AED 30,000 a month · where the answer is less

Thirty-two cells planned, two affordable. The correct recommendation is one market and one platform, which makes you look less ambitious than the planner who says yes, and is right anyway.

Honest limits

What is not in the box.

Stated here rather than discovered after you buy.

  • No current CPM, CTR or conversion-rate figures. The planning ranges ship with their eleven columns and without numbers. Filling them with plausible averages would make a guess look like a default, which is the exact failure this system is against. You fill them from your own accounts.
  • No universal platform minimum spends. They are objective-dependent and change without announcement. Rather than invent one, the course teaches you the three honest ways to work out the minimum for your own objective, and where to confirm it.
  • No client deck template yet. The Client Export tab produces the content; the designed PPTX is not in this edition.
  • No video, no login, no app. XLSX and PDF. Nothing that can disappear.
  • No updates. 2026 Edition is a dated snapshot. Every dated claim carries its own source and both dates so you can see exactly what will age.
Where no defensible number exists, you get the method instead of a guess. A course that shows you the edges of what is knowable is easier to trust than one that appears to have none.