PPC BudgetingWorking backwards from a customer

The wrong question and the right one#

Most budget conversations start with 'how much should we spend on ads?' and end with a round number based on last year, a percentage of revenue or what feels comfortable. None of these say anything about whether the money will come back.

The better question is 'how much can we profitably spend to acquire a customer, and how many such customers can the market supply at that price?' The budget is the answer to that question, not an input to it. It may turn out to be larger than you expected, or smaller, but it will be defensible.

Step one: what can you afford per customer?#

Start with the economics of a customer. What is the gross margin on a first order? What do customers typically spend over a year, or over their lifetime? How much of that margin are you willing to invest to acquire them? The answer is your maximum allowable cost per acquisition, sometimes called target CAC.

Be honest about which value you use. Basing allowable cost on lifetime value lets you spend more, but only works if you have the cash to wait for repeat purchases and evidence that they happen. Basing it on first-order margin is safer and slower. Our guide to CAC and LTV explains the trade-off.

Step two: how many customers do you need?#

From the business plan, work out how many new customers paid media needs to deliver in the period. Then work backwards through the funnel to the clicks needed, using your own conversion rates where you have them and conservative assumptions where you do not.

If the implied cost per customer exceeds what you can afford, more budget will not help. You need a better conversion rate, a higher-value offer, a cheaper channel, or a smaller target. This is the most useful moment in budget planning, because it forces the conversation about what must change.

Step three: is there enough demand?#

Search budgets are bounded by how many people search. Keyword planning tools give rough estimates of search volume and cost; treat them as directional, not exact. If your targets require more clicks than the relevant searches can supply, you will need to widen keywords, add other channels or accept a smaller target.

Social and video platforms are less constrained by volume, but costs rise as you reach beyond the most responsive audiences. In both cases, the last rupee spent is usually less efficient than the first. Budget decisions should be made at the margin: what does the next increment buy?

Step four: size the learning budget#

A new channel or campaign needs enough spend to produce a decision. If a test budget buys only a handful of conversions, you cannot tell a good campaign from a lucky one. Size test budgets by the number of conversions you need to see, not by what feels affordable to lose.

A simple approach: decide how many conversions would give you confidence, multiply by a realistic cost per conversion, and treat that as the minimum test budget. If you cannot afford that, test fewer things, or optimise toward an earlier funnel event that happens more often, while recognising its weaker link to revenue.

Step five: allocate and pace#

Plan seasonality explicitly. Demand in many categories rises around festive seasons, sales events or financial year ends, and auction costs often rise with it. A flat monthly budget will underspend in peaks and overspend in troughs. Shape the budget to demand, and use platform tools for short-term seasonal adjustments where appropriate.

Daily budgets on most platforms are averages, not hard caps. Platforms may spend more on some days and less on others while staying within a monthly limit. Check current platform rules so pacing does not surprise your finance team.

When to scale#

Scale when a campaign is profitable at the margin, not just on average. A campaign with an excellent average cost per acquisition may already be spending into its least efficient auctions. Increase budgets in steps, watch what the additional spend buys, and stop increasing when marginal cost per acquisition approaches your allowable limit.

Budgeting for a new account#

Without history, every figure in the plan is an assumption. That is not a reason to avoid planning; it is a reason to make the assumptions explicit and to design the first months to replace them with evidence. Write down the cost per click, conversion rate and close rate you are assuming, and where each came from.

Then treat the first phase as a measurement exercise. Spend enough to replace the riskiest assumption with a real number, usually conversion rate or close rate, before committing to a full-year plan. A business that discovers in month two that its close rate is half what it assumed has saved ten months of planning on a false base.

Finance teams often prefer a fixed number for the year. A useful compromise is a committed base budget for proven activity, a learning budget with a fixed ceiling, and a scaling reserve released only when agreed thresholds are met. This gives finance predictability and gives marketing room to move quickly when something works.

Budgeting mistakes to avoid#

• Spreading thin. Many small campaigns each too small to learn.
• Judging on averages. Ignoring the declining return of each extra rupee.
• Cutting prospecting to fund retargeting because retargeting reports a better return.
• Ignoring the sales cycle. Judging long-cycle B2B spend after a few weeks.
• Treating budgets as fixed for the year when demand and performance change monthly.

Paid media budgets sit inside the wider marketing budget, and the split between channels belongs in a media plan. For the measures you will steer by once money is flowing, see ROAS versus CPA.

Published by Fabulous.Media, a network of specialist marketing agencies. Updated 9 October 2026. Platform features change often; check current official documentation before acting on platform-specific detail.