Strategy & Craft
Splitting an ad budget: the 70-20-10 rule
Seventy per cent on what works, twenty on scaling, ten on experiments. The rule is old, simple, and violated in almost every account.
2 min readMay 9, 2026
Strategy & Craft
The metrics that actually count
Most dashboards track twenty numbers and steer by none. Five figures cover almost every advertising decision a mid-sized firm makes.
Key points
- Any metric you would not change a decision on does not belong in the report.
- Cost per enquiry without a close rate is meaningless.
- Track the same five numbers for three years and the pattern does the work.
Reporting tends to grow. Each new tool adds numbers, nothing is ever removed, and eventually nobody reads the report because it takes forty minutes and changes nothing.
The five that matter
1. Enquiries per week, total. All sources combined. This is the number the business feels.
2. Cost per enquiry. Total advertising spend divided by total enquiries. Not per channel; per business. Channel-level figures are diagnostics.
3. Enquiry-to-customer rate. What proportion of enquiries becomes an order. This is where advertising and sales meet, and where most improvement is available.
4. Cost per customer. Metrics two and three combined. The number that decides whether advertising is profitable.
5. Average customer value. Ideally over the full relationship, not the first order.
Metrics four and five together answer the only question that matters: does a customer cost less than they are worth?
A cost per enquiry of £40 sounds excellent until you learn that only 8 per cent of those enquiries buy. The real figure is £500 per customer.
Why per-channel cost per enquiry misleads
Channels produce different enquiry quality. A cheap channel delivering enquiries that never close is more expensive than a costly one delivering enquiries that do.
The fix is to track the close rate by source, which requires recording the source on every enquiry and carrying it through to the order. That is a process change, not a tool purchase, and it is the highest-value reporting improvement most firms can make.
The diagnostics tier
These are not objectives. They explain movements in the five.
| Diagnostic | Explains |
|---|---|
| Impressions and reach | why enquiries rose or fell |
| Click rate | whether the message is landing |
| Landing page conversion | whether the page is the bottleneck |
| Response time to enquiries | why close rate changed |
| Average order value | why revenue moved without volume moving |
The metrics to stop reporting
Followers, likes, page views, bounce rate, time on page, and any platform's own "engagement score". None of them changes a decision, and each one absorbs attention that belongs elsewhere.
If someone objects, ask what they would do differently at a 10 per cent change in the number. If there is no answer, it is not a metric, it is a hobby.
The three-year view
Single months are noise. The value of a small, stable metric set is that after three years you have a chart that shows the actual trajectory of the business, with the seasonal pattern visible and the effect of major changes readable.
That chart is worth more than any dashboard, and it only exists if you resist changing what you measure.
Frequently asked questions
What about impressions and reach?
Useful as diagnostics when something changes, useless as objectives. Nobody has ever banked an impression.
How often should they be reviewed?
Weekly for spend and enquiries, monthly for cost per customer, quarterly for the rest.
More from Strategy & Craft
Keep reading
Strategy & Craft
How much advertising budget is the right amount
Percentage-of-revenue rules are widely quoted and mostly wrong. The answer comes from your own numbers in four steps.
2 min readMay 7, 2026
Strategy & Craft
The brief agencies actually need
A bad brief produces bad work at full price, and the client pays twice. Eight questions, two pages, three hours of your time.
2 min readMay 5, 2026