Formula and what the number really is
CPC = campaign spend / number of clicks
Example: $1,200 / 2,000 clicks = $0.60
CPC is not a rate card, it is the result of an auction. The advertiser sets a maximum bid, but the platform charges the amount required to hold the position relative to the nearest competitor. That is why the actual CPC is almost always below the declared maximum, and why “lower the bid” is not the same as “lower the CPC”: below a certain level the ad simply stops being shown.
| Factor | How it moves CPC |
|---|---|
| Competition for the query | The main driver: the more high-margin advertisers bid, the dearer the click |
| Ad quality and relevance | A high CTR and a matching landing page lower the charged price at the same position |
| Product category | Expensive goods with a long decision cycle cost more per click |
| Seasonality | Demand peaks lift bids across the whole category at once |
| Query type | Brand queries are cheaper, broad category queries dearer, transactional dearer than informational |
| Device and region | Different auctions with different competitive density |
The chain: CPC → CR → CPO → ad cost ratio
On its own CPC says nothing about efficiency. It only becomes manageable inside the chain that runs to money:
CPO = CPC / CR
Ad cost ratio = CPO / AOV × 100%
Take it through the numbers. The starting campaign:
CPC = $0.60
CR = 1.0%
AOV = $120
CPO = 0.60 / 0.01 = $60
Ad cost ratio = 60 / 120 × 100% = 50%
Fifty per cent is a knowingly loss-making level for most categories. Now double the landing-page conversion rate without touching the ad account at all:
CPC = $0.60 (unchanged)
CR = 2.0%
AOV = $120
CPO = 0.60 / 0.02 = $30
Ad cost ratio = 30 / 120 × 100% = 25%
Cost per order halved, and so did the ad cost ratio — at an unchanged click price and an unchanged volume of traffic.
Why fighting over CPC pays less than working on conversion
Compare three optimisation scenarios from one baseline (CPC $0.60, CR 1.0%, AOV $120, budget $6,000).
| Scenario | CPC | CR | Clicks | Orders | CPO | Ad cost ratio |
|---|---|---|---|---|---|---|
| Baseline | $0.60 | 1.0% | 10,000 | 100 | $60 | 50% |
| CPC −20% (paid for in positions) | $0.48 | 0.9% | 12,500 | 113 | $53 | 44% |
| CR +20% | $0.60 | 1.2% | 10,000 | 120 | $50 | 42% |
| CR +20% and AOV +10% | $0.60 | 1.2% | 10,000 | 120 | $50 | 38% |
The figures are illustrative, but the relationship holds for two reasons:
- Cutting CPC is rarely free. Moving to cheaper positions and queries usually brings traffic with weaker intent — conversion falls along with the click price, and part of the saving is eaten up.
- Conversion growth has no ceiling on the auction side. It is limited only by the quality of the user’s path after the click, and its effect spreads across all traffic, not just the paid part.
There is a third reason, an organisational one. Bids are owned by the traffic specialist, post-click conversion by the product and e-commerce teams. Because of that split, optimisation defaults to the first half of the chain: it is closer to hand, easier to reason about and measured in the same account where the money is spent. The second half — how well the landing page keeps the promise of the ad, how fast a product can be chosen, how good the selections are — often never enters the conversation about advertising efficiency at all, even though it sits in the denominator of the same formula.
A useful habit is to count not CPC but the “price of a useful click”: spend divided by the number of clicks that reached at least the product page or a second screen. A campaign at $0.48 per click with 40% instant bounces is more expensive than a campaign at $0.60 per click with 15%.
How to work out your maximum CPC
The acceptable click price follows from your economics, not from benchmarks:
Maximum CPC = AOV × margin × CR
Example: $100 × 30% × 1.5% = $0.45
Anything above that threshold loses money on the first purchase. You can work above it deliberately — if lifetime value is known and the customer comes back — but then the decision is made explicitly, with a payback horizon attached.
The formula also reads the other way round: there are three ways to buy more traffic without breaking the economics — raise the average order value, raise the conversion rate, or raise basket margin through add-on products. Each of them lifts the ceiling on acceptable CPC and lets you buy dearer, better traffic while staying inside your target revenue per visitor.