Ad Budget Forecaster

What's a realistic monthly growth rate for ad spend?

As a rule of thumb, 10-20% a month once an offer is proven. That is sustained growth, not step size: a single budget step is often about 20%, but each needs time to settle and prove itself before the next. Much faster scaling (30%+ a month) tends to push ad sets back into learning and reach less-interested people, so cost per lead rises. Use the ROAS-drop input to see how that erodes the projection.

Should I use my actual ROAS or a target?

Target — slightly below your current ROAS to stay conservative. Scaling usually pushes ROAS down as you reach less-qualified audiences. Plan for a lower ROAS than month 1 showed so you're not disappointed.

When should I stop scaling?

When marginal ROAS drops below your profitability threshold. Stop when the ROAS on the last dollar falls to about 2 ÷ margin (5x at a 40% margin), because beyond that each extra dollar takes more than half the gross profit it wins. The tool shows that marginal ROAS and the month where monthly gross profit after ads peaks; scaling past the peak lowers profit. Not every contractor should scale to $10K/mo; smaller markets can cap out at a few thousand dollars a month.

Does this account for seasonality?

No — this assumes steady compound monthly growth. In reality, home service trades have seasonal peaks (roofing in hail season, HVAC in summer). Use this for the overall trend, then layer the monthly multipliers from the Seasonality Calendar (/tools/seasonality-calendar) on top for campaign planning.

What target ROAS should I use for projections?

Start with your actual last-90-day ROAS, minus 10-15% safety margin. Reasoning: ROAS usually drifts lower as spend scales into less-qualified audiences. Hypothetically, if your last 90 days averaged 4.5x, projecting at 3.8-4.0x (a 10-15% haircut, a rule of thumb) keeps the forecast honest. Better to beat a conservative forecast than miss an optimistic one.

When should I plan to revisit and adjust the forecast?

Every 30 days. Compare actual vs projected on three numbers: ad spend (was your scaling pace right?), revenue (did ROAS match the model?), close rate. If actuals trail projection by a wide margin (say 20%+, a rule of thumb), you have either a market saturation issue (slow scaling further) or an internal funnel issue (use the Close Rate Calculator to diagnose). Models that don't get revisited turn into pretty fiction.

Should I share this projection with my financing/lending partner?

Yes — but with appropriate caveats. Lenders look at projections to gauge cash flow. Show them: (1) the conservative version (for example 15-20% below actuals) and label it 'conservative'; (2) the realistic version labeled 'expected'; (3) NEVER share the optimistic version alone. Lenders trust contractors who show range and discount confidence; they distrust contractors who promise hockey-stick growth with no plan B.

How should I budget for creative production alongside ad spend?

A common rule of thumb is to reserve 15-25% of your monthly ad budget for creative production: at $3K/mo ad spend, that's $450-750/mo. Too little and stale creative drags performance; too much and production outruns what the budget can test. Fresh creative is what keeps fatigue from pushing cost per lead up. Creative production isn't an expense — it's an asset that depreciates if not refreshed. Build the creative budget into the forecast from month 1, not as an afterthought when ads start fatiguing.

What growth rate is realistic in year 2 vs year 1 of running Meta ads?

In year 1, grow spend in steady steps — around 10-20% a month is a common rule of thumb once an offer is proven, because bigger jumps tend to push the account back into learning and raise cost per lead. Expect growth to slow in year 2, with longer plateaus and more focus on efficiency than scale. The math: compounding 30% a month takes you from $1K in month 1 to about $18K by month 12, far more than most local markets can absorb before the same people see your ads over and over. Year 2 growth comes from: expanding service area, adding service lines, layering new platforms (Google LSA, YouTube Shorts), not from cranking the same Meta lever harder. Models that assume year 1 growth rates continuing forever produce wildly inflated revenue projections — and disappointed contractors. Build the deceleration into your forecast.

Should the forecast model include ad-spend cuts during slow weeks (holiday weeks, weather disruptions, vacation weeks)?

Yes — but only for predictable slow windows. Three windows worth modeling cuts: (1) HOLIDAY WEEKS — Thanksgiving week + the 7-10 days between Christmas + New Year. Consider cutting spend sharply; booking activity is usually much lower in these weeks for non-emergency work; (2) MAJOR LOCAL EVENTS — Super Bowl Sunday, election week, school graduations. attention is elsewhere, so a short, modest cut for those few days is reasonable; (3) PLANNED VACATIONS — when YOUR sales team is at half-capacity. No point generating leads if no one's available to follow up within 1 hour. Build these cuts into the annual forecast, then check your own booking data to confirm the slow windows are really slow in your market. Running a flat budget through the holidays spends money on impressions that rarely convert. Forecast smarter than 'same spend every week.'

How do I model the impact of adding a second ad platform (Google LSA or YouTube) on top of my Meta budget?

Two-step approach. First, DON'T assume new platforms compound — they often cannibalize. A customer who would have come through Meta might now come through LSA at the same revenue but a different cost. Second, model platforms separately + then run a 90-day attribution audit to confirm true incremental contribution. Realistic expectations: (1) ADDING LSA — early numbers overstate its contribution, because some of those jobs would have come through Meta or organic search anyway; (2) ADDING YOUTUBE — expect more overlap with Meta, because both reach discovery-stage audiences. Only the before-and-after comparison below tells you the true incremental figure for your business. Budget rule: when adding a new platform, hold Meta spend FLAT (don't increase) for the first 90 days. Compare total business growth in that window vs the 90 days prior. The delta is your true incremental from the new platform — not whatever it's reporting.

What's the right way to model cash flow vs ad spend when revenue lags by 30-90 days?

Model two parallel cash views: a SPEND-OUTFLOW timeline + a REVENUE-INFLOW timeline, offset by your trade's typical payment delay. Three steps: (1) MAP your typical lead-to-payment delay — rough planning ranges: emergency services, days (pay on completion); roofing/HVAC installs, a few weeks (deposit + final invoice); remodeling, one to three months (multiple payment milestones). This is your 'lag'; (2) FRONT-LOAD the spend timeline — month 1 ad spend = $5K outflow, but month 1 revenue is from leads acquired BEFORE you started this campaign. Real revenue from month 1 ad spend doesn't hit until month 2-4 depending on your trade; (3) BUILD WORKING CAPITAL — keep a cash reserve of, as a rule of thumb, 1.5-3x your monthly ad spend so you don't lose campaigns to cash flow gaps. Pausing Meta 'because cash is tight' often happens at the worst possible time — the spend has already paid for leads still in your sales funnel. Pausing kills future revenue without saving past costs. Working capital prevents the panic-pause.

What's the right way to incorporate seasonality into a 12-month forecast vs treating each month independently?

Two-tier model. TIER 1 (seasonality multipliers): use the Seasonality Calendar (/tools/seasonality-calendar) to set a per-month multiplier vs your 'baseline' month. Hypothetical roofing example: April-June = 1.4x baseline; Jan-Feb = 0.5x baseline; March/Sept-Oct = 1.0x. Apply the multiplier to your baseline budget to get monthly forecast. TIER 2 (CPL adjustments): peak months usually cost more per lead too, as competitors bid for the same homeowners. Check your own history for the size of the premium; if you have none, assume something like 10%. Combined model: April spend = baseline × 1.4 (volume) × 1.10 (CPL premium) = 1.54x the baseline budget to buy 1.4x the baseline leads. Flat monthly forecasts lead to panic when peak months over-burn or off-season months under-deliver. Two-tier model = realistic 12-month outlook. Use this forecaster for the overall growth path, then layer the monthly multipliers and CPL premium on top.

How do I forecast at multiple confidence levels (best case / expected / worst case) and which one drives my actual budget decisions?

Build all three; budget against EXPECTED. In this tool, that means three runs: (1) BEST CASE — starting ROAS from your best month and a small ROAS drop; use it only for stretch goals; (2) EXPECTED — starting ROAS from your last 90 days and the default 10% drop; this is your decision-making baseline; (3) WORST CASE — starting ROAS from your worst month, minus about 10% (a rule of thumb), with a steeper drop such as 20%; this is your downside risk envelope. Budget rule: PLAN against EXPECTED. CASH-RESERVE against WORST CASE. Stretch goals from BEST CASE inform creative + offer experimentation budget but never the operating plan. It's easy to err in two directions: (a) plan against best-case → cash-flow disaster when reality hits expected; (b) plan against worst-case → never invest in growth opportunities. The middle path keeps you calm during volatility + capable of seizing scaling opportunities. Run all three quarterly; document them. Decision-makers should know all three numbers at any given moment.

How does the forecast change if I anticipate hiring an agency mid-year vs running DIY all year?

An agency hire should improve results after a ramp period — but it adds management fees, so model it rather than assume it pays. Three scenario adjustments: (1) PRE-AGENCY months (DIY phase) — use your current actual CPL + close rate as inputs; this is realistic; (2) TRANSITION months (agency onboarding, weeks 1-8) — assume FLAT performance vs DIY (no improvement yet); (3) POST-RAMP months (agency month 3+) — adjust inputs to what the agency commits to in writing, not to a hoped-for improvement. The hidden math: agency fees have to be offset by performance lifts. If the hire isn't paying for itself a few months after ramp, raise it with the agency. Build the forecast with both scenarios — DIY-only year + agency-mid-year. Compare projected total revenue minus all costs. The decision becomes math-grounded vs feeling-grounded. It's tempting to decide on 'I don't have time anymore' alone; run both scenarios and let the numbers decide.

What's the right way to use this forecaster to set a year-end revenue target — and is it realistic to set 'aggressive' targets?

Compute three scenarios; commit to EXPECTED publicly + STRETCH privately. Three-scenario template: (1) CONSERVATIVE = current trailing-90-day CPL + close rate held constant + flat budget = baseline revenue projection; (2) EXPECTED = same metrics with a modest improvement through the year (say 8-12%, a rule of thumb) + planned budget growth; (3) STRETCH = bigger improvements (say 15-25%) + accelerated budget growth. Public commit (to lenders, investors, team): conservative or expected — never stretch. Internal use: stretch targets motivate effort + identify upside scenarios. The math: aggressive forecasts that miss create credibility damage; conservative forecasts that over-deliver build trust. One optimistic forecast committed to publicly sets up disappointment and erodes trust. The 3-scenario approach lets you set realistic external expectations + still chase upside internally. By Q4, calibrate next year's forecast based on which scenario your actual numbers tracked closest to.