What it means
Banner adverts are the oldest form of paid digital media and they still account for a large share of online advertising budgets. They appear in standard sizes across publisher sites, are sold through advertising exchanges in fractions of a second, and are targeted using signals such as the page content, the visitor's location or previous visits to the advertiser's own site.
The commercial logic is different from search advertising. Someone typing a query into a search engine has already declared an intention to buy, whereas a banner interrupts someone reading about something else, so click rates are low and are frequently well below 1%.
That does not automatically make banners poor value, but it does mean judging them on clicks alone will always flatter search and punish display. Pricing normally uses cost per thousand impressions, written as CPM, where M is the Roman numeral for a thousand.
Some campaigns are bought on cost per click or cost per acquisition instead, which shifts risk to the publisher or network and usually carries a higher effective price. Finance teams should recognise that the pricing model changes who carries the risk of a campaign that fails to perform.
The main disputes in this area are about measurement rather than creative work. Viewability standards define whether an advert that loaded below the fold really counts as seen, attribution rules decide whether a banner viewed but not clicked should receive credit for a later purchase, and both choices can move the reported return on a campaign by a wide margin.
In practice
Real-world examples.
Example
A regional insurance company buys banner space on local news sites at a $4 CPM, mainly so that householders recognise its name when a broker mentions it. It measures success by brand search volume rather than clicks, and sees searches for its name rise 18% during the campaign months.
Example
An online furniture retailer runs retargeting banners aimed only at visitors who left items in a basket in the previous 14 days. The click-through rate reaches 0.6%, roughly eight times its general display average, and the cost per acquisition falls to a level the finance team is comfortable funding year round.
Example
A business software vendor pays a specialist industry publisher a flat $15,000 a quarter for a fixed banner position on its pricing comparison page. The value is the placement itself, so the deal is negotiated on position and exclusivity rather than on impressions delivered.
Formula
Calculation
Spend = (impressions / 1,000) x CPM. Clicks = impressions x click-through rate. Cost per click = spend / clicks. Cost per acquisition = spend / conversions.
A subscription software company runs a banner campaign delivering 2,000,000 impressions at a CPM of $6.00, with a click-through rate of 0.08% and a landing page conversion rate of 2.5%.
Spend: (2,000,000 / 1,000) x $6.00 = 2,000 x $6.00 = $12,000.
Clicks: 2,000,000 x 0.08% = 1,600 clicks.
Cost per click: $12,000 / 1,600 = $7.50.
Conversions: 1,600 x 2.5% = 40 sign-ups.
Cost per acquisition: $12,000 / 40 = $300.
If each new subscriber is worth $900 in gross profit over their expected life, the campaign returns $900 / $300 = $3 of profit for every $1 spent, which is the number that belongs in the board pack rather than the click-through rate.Case study
Seen in the real world.
Meridian Kitchenware is an illustrative, fictional homewares brand that spent $12,000 a month on banner advertising and could not agree internally on whether it worked. The marketing team pointed to 2,000,000 monthly impressions and rising brand awareness, while the finance team pointed to a cost per click of $7.50 against $1.10 on search and wanted the budget cut.
The chief financial officer settled the argument with a test rather than an opinion. Banner spend was switched off in three regions for eight weeks while search spend was held constant, and total revenue in the switched-off regions fell about 9% relative to the control regions, more than the direct banner-attributed sales had suggested.
In this fictional case the conclusion was to keep the display budget but move most of it into retargeting, where the audience was already familiar with the brand. The point of the story is the method: for advertising that mainly influences rather than converts, a holdout test is worth more than any attribution model.
Watch out
Common mistakes.
- Judging banner campaigns purely on click-through rate. Most of the value of display advertising is in views that never produce a click but do influence a later purchase.
- Paying for impressions without checking viewability. An advert that loads far below the visible area of a page can be counted and billed even though nobody ever saw it.
- Assuming last-click attribution is neutral. It systematically credits the final search advert and hides the display activity that created the demand in the first place.
Questions
People also ask.
What does CPM actually mean?
It is the price for one thousand impressions, so a $6 CPM means each individual advert view costs $0.006.
Are banners still worth buying when so many people use ad blockers?
Yes for many advertisers, because blocked impressions are generally not billed and app and connected television inventory is far less affected.
How do I prove display advertising works?
Run a geographic holdout test, switching the spend off in comparable regions and comparing total revenue rather than relying on tracked clicks.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related