Three questions, three answers, then the workings.
1. Is Google in the forecast? Yes, and it has been since 15 August. Every spend figure in the August tracker and in this document is Meta plus Google, pulled per day from each platform. What is not right is the assumption underneath the old plan: it treated Google as a steady £230/day. Google has drifted up to £320/day. More importantly, Google spend is not a lever we control. There is £618/day of budget sitting against about £320/day of actual spend, and every campaign reports 0.0% impression share lost to budget. Raising Google budgets would buy nothing.
2. What does the budget cut do to August? Less than it looks. Meta configured budget is now £1,240/day, down from about £1,446. But Google has drifted up by roughly £80/day at the same time, so the net change is about £142/day. August lands at £52,270 of ad spend instead of £53,828. The revenue cost is around £2,571 of new-customer revenue across the remaining 11 days.
3. Is the September goal reachable, and is it profitable? Reachable, but only by roughly doubling ad spend to about £3,029/day (£90,870 for the month). And here is the finding that actually matters: in-month profit is almost identical at every spend level we modelled, between £31,735 and £34,619. Spending nearly twice as much changes September profit by about £2,884. So this is not a profitability decision. It is a cash and lifetime-value decision, and it should be argued on those terms.
The client goal is £268,147, split £58,897 subscription and £209,250 everything else. Here is that number against three reads built from data rather than from a target.
| Read | September total | vs goal | Basis |
|---|---|---|---|
| Client goal | £268,147 | the target | What Ailis has asked for. |
| Current growth rate carried forward | £209,292 | −£58,855 | September 2025 (£106,024) times the 1.97x year on year the business is actually running at. |
| Bottom-up at today’s budget | £197,902 | −£70,245 | Subscription ledger, plus returning revenue at run rate, plus what the post-cut budget buys. |
| Bottom-up at the budget needed to hit goal | £268,143 | −£4 | Same build, with ad spend at £3,029/day. |
The business is growing fast. Comparing the same window in both years, 1 to 20 August, revenue went from £63,688 to £125,719. That is 1.97x year on year. Carry that rate onto last September and you land near £209,292. The client goal is 28% above the rate the business is currently compounding at, and 2.53x last September. It is not a fantasy number, but it does not happen by carrying on as we are.
Revenue splits into three pieces that behave completely differently. Only one of them responds to ad spend, and sizing the budget against the other two is the mistake that wastes money.
| Component | September | How it is forecast |
|---|---|---|
| Recurring subscriptions | £64,207 | Exact forward ledger from Recharge. Not a statistical estimate. |
| Returning customers, non-subscription | £54,000 | £1,800/day run rate. Jun £52,294, Jul £46,930, Aug tracking similar. Driven by email and calendar, not ads. |
| New customers | the balance | This is the only line ad spend moves. Everything above is fixed before we set a budget. |
| To reach the £268,147 goal, new-customer revenue must be | £149,940 | That is 76% more than August's run rate of £85,385, and needs about 2,068 new customers at the current £73 new-customer AOV. |
In August the client's subscription target was the single biggest source of error. It is worth being precise about why, because the September answer is the opposite.
Until now the subscription forecast was inferred from billing history, because the Shopify API refused access to subscription contracts. That block is gone. Pott's Recharge account is directly readable, so we now have the actual forward charge ledger: 3,070 active subscriptions, every one with its next charge date and amount. This is a schedule, not a model.
| Subscription revenue | August | September |
|---|---|---|
| Client's plan | £58,265 | £58,897 |
| Recharge forward ledger | £48,286 | £64,207 |
| Variance vs plan | −£9,979 (-17.1%) | +£5,310 (+9.0%) |
August misses her subscription target by about £9,979. September beats it by about £5,310. That is not luck. August was an unusually strong month for new subscription sign-ups, £17,615 in the first 20 days against roughly £14,000 in a normal month, and a new subscriber's first recurring charge lands about a month later. September is collecting August's sign-ups. £53,269 of September is already sitting in the Recharge queue with a date and an amount against it; the remaining £15,771 is late-August billers rolling round again.
Range on that number is £62,136 to £66,969 depending on how many charges skip or fail. We have used £64,207. Worth confirming with Ailis that her £58,897 means recurring charges only, and does not include first-time subscription orders, which land in new-customer revenue and are running £17,615 per 20 days on top.
All four scenarios hold new-customer ROAS at 1.65, which is what August has run at (1.65) and what last September ran at (1.66). Profit is contribution at 56.5% less ad spend less £28,896 of monthly overhead.
| Scenario | Spend/day | Month spend | New customers | Revenue | % of goal | In-month profit | 36-month LTV bought |
|---|---|---|---|---|---|---|---|
| Hold post-cut budget | £1,610 | £48,300 | 1,099 | £197,902 | 74% | £34,619 | £113,398 |
| Restore pre-cut budget | £1,750 | £52,500 | 1,195 | £204,832 | 76% | £34,334 | £123,259 |
| Staged ramp to £2,500 | £2,200 | £66,000 | 1,502 | £227,107 | 85% | £33,420 | £154,954 |
| Goal-chasing | £3,029 | £90,870 | 2,068 | £268,143 | 100% | £31,735 | £213,344 |
Read the profit column and the LTV column together. Going from £1,610/day to £3,029/day costs £42,570 more in cash and reduces September profit by £2,884, which is noise. In exchange it buys 969 additional new customers worth about £99,946 of contribution over 36 months.
The reason profit barely moves is arithmetic. New-customer ROAS of 1.65 at a 56.5% margin returns 0.93 of contribution per pound spent. Break-even inside the month is 1.77 NC ROAS. We are just under it, so every extra pound is very slightly dilutive in September and clearly accretive over the customer's life. September profit comes from the £118,207 of subscription and returning revenue that carries almost no marginal ad cost.
The whole goal case rests on new-customer ROAS holding at 1.65 while spend nearly doubles. Here is what happens if it does not.
| At £3,029/day, if NC ROAS is | CAC | Revenue | % of goal | In-month profit |
|---|---|---|---|---|
| 1.85 | £39 | £286,317 | 107% | £42,003 |
| 1.65 today | £44 | £268,143 | 100% | £31,735 |
| 1.45 | £50 | £249,969 | 93% | £21,466 |
| 1.30 | £56 | £236,338 | 88% | £13,765 |
Evidence that it can hold. Cost per new customer is flat across every spend level we have observed this year. Sorting the last 112 trading days into spend quartiles, the account ran £944/day to £1,643/day with CAC moving only between £44 and £49. There is no saturation in that range. Better still, in November 2025 this account spent £102,810 in a single month at a new-customer ROAS of 1.91, so it has absorbed goal-level spend before without breaking.
Evidence for caution. November is peak season and September is not. Last September the account spent £29,782 in total. We would be asking the same seasonal window to absorb 3.1x that. We have no in-season evidence above roughly £1,900/day, so everything past that point is extrapolation.
Do not jump to £3,029/day on 1 September, and do not sit at £1,610/day either. Ramp, with a hard guardrail.
| Window | Meta/day | Total/day | Condition to move to the next step |
|---|---|---|---|
| 1 to 7 Sept | £1,700 | £2,020 | Restore the cut, plus a step. Move on if 7-day CAC stays under £55. |
| 8 to 14 Sept | £2,100 | £2,420 | Same test. This is already past our observed evidence, so watch it daily. |
| 15 to 22 Sept | £2,500 | £2,820 | Same test. Hold here if CAC drifts over £55. |
| 23 to 30 Sept | £2,900 | £3,220 | Only if CAC held. This also warms the account up for October. |
| Month | £2,647/day avg | £79,400 of ad spend across 30 days. |
That path spends £79,400, an average of £2,647/day, and lands near £249,217 if ROAS holds, which is 93% of the goal. Going the rest of the way is possible but should be a deliberate decision taken in the second week, once the ramp has produced real evidence at £2,420/day rather than an assumption.
The trigger to stop is sustained CAC above £55, not a falling blended ROAS. Blended ROAS will keep sliding while we scale, because spend is growing against a £118,207 base of subscription and returning revenue that ads do not touch. That slide is arithmetic, not a leak, and reacting to it would cut the wrong thing.
Three things to settle before 1 September. One, confirm with Ailis that £58,897 means recurring charges only. Two, the cash question is real: £90,870 of September spend against Q4 inventory commitments needs her sign-off, because the return on it arrives in month four, not in September. Three, Google needs attention on its own terms, since it is running £320/day against £618/day of available budget with no budget-limited impressions. Extra Google volume has to come from bids, tROAS targets or the feed, and none of the scaling above assumes any of it.
The budget change on 19 and 20 August, and what August now lands at.
| Meta campaign | Daily budget | Last changed |
|---|---|---|
| CS | Sales | Cold & Warm | Evergreen (old Tomato vine) | £800 | 2026-08-20 |
| CS | Sales | Cold & Warm | Roses May 2026 | £130 | 2026-08-19 |
| CS | Sales | Cold & Warm | Citronella May 2026 | £100 | 2026-08-19 |
| CS | Sales | Cold & Warm | Cook Candles | £80 | 2026-08-04 |
| CS | Testing | Jul 2026 > Test - Tomato Vine - Broad - Exclude Customers | £50 | 2026-07-25 |
| CS | Testing | Jul 2026 > Test - Cook Candle - Broad - Exclude Customers | £50 | 2026-07-25 |
| CS | Sales | Hot | Retention | Subscription Opt-In | 7D Click 1D View | £30 | 2026-07-14 |
| Total configured | £1,240 | was about £1,446 |
| August | Old plan (flat £1,750/day) | Restated |
|---|---|---|
| Spend, 21 to 31 Aug | £19,250 | £17,692 |
| Month ad spend | £53,828 | £52,270 |
| Subscription revenue | £53,063 (frozen model) | £48,286 (ledger) |
| Total revenue, run rate | n/a | £194,182 |
| Total revenue, if her online plan lands | n/a | £208,413 |
| Her full-month target | n/a | £225,695 |
Her plan for 21 to 31 August assumes £66,000 of non-subscription revenue, against a run rate of £51,770. That is a 27% lift, resting on the bank holiday push on the 25th and the Tomato Vine dates. Possible, but it is the optimistic end.
Worth being explicit, because this was the original question.
| Campaign | Daily budget | Actual/day (13-19 Aug) | Budget-lost impr. share |
|---|---|---|---|
| CS | Shopping | High | All Products | £264 | £192 | 0.0% |
| CS | Shopping | Low | All Products | Brand | £264 | £52 | 0.0% |
| CS | Search | Brand | £45 | £25 | 0.0% |
| CS | Performance Max | All Products | £45 | £24 | 0.3% |
| Total | £618 | £293 | none |
Every campaign is spending well under budget with no impressions lost to budget. Google is finding as much demand as it can at current bids and is limited by the auction, not by us. It is forecast in this plan at £320/day, treated as a floating input we predict rather than a number we set. That is the opposite of Meta, where the budget is a decision.
| Input | Value | Source and confidence |
|---|---|---|
| Contribution margin | 56.5% | Weakest input. KC constant, still under the cost audit. Duplicate-SKU cost conflicts in Shopify are unresolved. At 50% September profit at goal spend falls to £14,305; at 60% it rises to £41,120. |
| New-customer ROAS | 1.65 | Aug 1-20 actual 1.65, Sept 2025 actual 1.66. Solid at current spend, unproven at £3,000/day in September. |
| Monthly overhead | £28,896 | KC client config. TW carries a stale £1,100/day entry from Oct 2025. Also part of the cost audit. |
| 36-month contribution per new customer | £103 | KC cohort curve. Payback lands in month 4 at a £44 CAC. |
| Charge realization | 93% | Share of queued Recharge charges that bill. Observed gross realization 93.1% / 91.8% / 89.2% for Jun / Jul / Aug, but that includes charges skipped well in advance which the live queue already excludes. Band 90% to 97%. |
| Returning non-subscription revenue | £1,800/day | Jun £1,743/day, Jul £1,514/day, Aug £1,874/day. Assumed flat, no September calendar supplied yet. |
| Meta spend vs configured budget | 1.039x | Observed 15-19 Aug: £1,503/day actual on £1,446/day configured. |
Not modelled: a September promotional calendar (none supplied), any October pull-forward, and any Google scaling. All three are upside to the numbers above rather than risk.