SEO vs Paid Ads for Ecommerce: Which Delivers Better ROI?
The generic version of this comparison does not fit a shop, because shopping advertising works off a product feed and behaves nothing like a text advertisement. That difference decides most of the answer.
The Core Difference
Advertising rents a position. Optimisation earns one. Rented position appears immediately and stops the moment payment stops. Earned position takes months to acquire and decays slowly rather than switching off.
What each buys. Two different things.
Advertising buys certainty and speed. You know what a click costs, you know when it starts. You can turn it off this afternoon. Optimisation buys an asset that keeps working, with no certainty about when it will arrive.
Why that framing is where most comparisons stop. And why stopping there is useless for a shop.
It applies equally to a solicitor, a plumber and a shop, which means it tells a store owner nothing about their own situation.
The three things that make a shop different. All covered below.
Shopping advertising runs off a product feed rather than off keywords and copy, which is block two. Margin sets a hard ceiling on what can be bid, which is block four. And a large catalogue produces a long tail that advertising covers expensively and optimisation covers cheaply, which is block seven.
What the answer usually is. Both, differently.
Most shops that do well run the two together. The question worth asking is what each is for, which block eight answers.
Shopping Advertising Changes The Comparison
On a commercial product search, the results a shopper sees first are frequently product listings with an image, a price and a shop name. Those are advertisements. They work in a way that changes what a shop is comparing.
Where they come from. A product feed rather than written advertisements.
The listings are generated from the shop's own product data: titles, descriptions, prices, availability and images. Nobody writes an advertisement for each product, which is what makes the format viable across a large catalogue at all.
Why that matters for the comparison. Three consequences.
The quality of your product data becomes an advertising input. The same fields that make a product page useful make the listing competitive, so improving them helps both channels at once.
They occupy the position a product page would want. On a search for a specific item, the paid listings sit where attention lands first, which is part of why our product page guide argues most product pages are not worth targeting.
They compete on price, visibly. A shopper sees several prices side by side before clicking anything, which suits a shop with keen pricing and punishes one competing on service or expertise.
What follows for a specialist. The searches where the shop's knowledge matters are the ones this format serves worst.
That is where optimisation has the clearer case, per block seven.
How Costs Behave As You Grow
The two behave completely differently as volume increases. That difference is the strongest argument for doing both rather than choosing.
Advertising scales with volume. Twice the orders costs roughly twice as much.
Efficiencies exist and are marginal against it. Cost per order stays broadly where it is, so growth does not make acquisition cheaper.
Optimisation does not scale that way. The cost stays flat while the volume it produces grows.
The same monthly fee that produced a small amount of traffic in month four produces considerably more in month twelve, because the work compounds rather than being consumed.
Where the crossover sits. Later than people hope. It does arrive.
Early on, advertising delivers and optimisation delivers little, so the comparison flatters advertising heavily. The lines cross at the point where accumulated optimisation is producing more than the fee would buy in advertising. On the shops we work on that is usually somewhere in the second year rather than the first.
What that means for a decision. A one year comparison will favour advertising. A three year comparison usually will not.
The caution attached to it. This only holds if the optimisation is working.
A flat fee producing nothing does not become good value through the passage of time. Our measurement guide covers how to tell which you have.
Margin Sets The Ceiling On Paid
A shop with thin margins runs out of room to bid long before a high margin one does. This is the constraint that decides how far advertising can take a particular business. It is arithmetic rather than strategy.
Why there is a ceiling at all. Because the gross profit on an order is the absolute maximum you can pay to win it.
Beyond that point you are buying revenue at a loss. Any target has to sit meaningfully below it to leave anything behind.
How to work out your own. Three steps, using figures you already hold.
Take your gross profit on a typical order, meaning what remains after cost of goods, fulfilment and returns. Decide what proportion of that you are willing to spend acquiring the order. Then divide that by the proportion of clicks that become orders, which gives the most you can pay for a click and still hit your target.
What that number tells you. Whether you can compete at all in your market.
If the figure you arrive at is below what clicks cost in your category, advertising cannot work at that margin however it is managed. That is not a failure of the advertising. It is the business model meeting the auction.
Why a competitor may seem to ignore this. Two reasons, both worth knowing.
They may have better margins than you, through volume buying or their own brand. Or they may be counting the lifetime of the customer rather than the first order, which our page on whether this is worth it covers.
What Happens When You Stop
This is the argument most often overstated in our industry, so it is worth being accurate about rather than persuasive.
Stop advertising. The traffic ends that day.
There is no residual effect on the paid channel itself. What was being bought stops being delivered, immediately and completely.
Stop optimisation. Visibility decays rather than disappearing.
The pages remain, the content remains and the structure remains. What stops is the maintenance. A shop is a moving target: new products arrive with supplier copy, apps get installed, competitors keep working and the platform changes underneath you.
How fast the decay is. Slower than suppliers imply and faster than clients hope.
A shop that stops usually holds its position for a while, then slips gradually as everything around it moves. The first thing to go is typically anything competitive, since that is where somebody else is actively working.
The claim we will not make. That optimisation is permanent.
It is not. It is durable, which is a different and smaller claim. The difference matters when somebody is deciding where to put money.
What genuinely persists. Structural work and content that answers something.
A category tree rebuilt around how customers search does not unbuild itself. A buying guide that helps somebody choose keeps helping. Those outlast the engagement in a way that a bid strategy cannot.
Where Paid Wins
Six cases where advertising is the correct answer and optimisation is not. We sell the second and these are still true.
A new shop with no history. The clearest case of the six.
A site nobody links to and nothing points at will take a long time to be found. Advertising works on day one, which can be the difference between a business surviving its first year and not.
A launch. Where something has to be visible on a specific date.
Optimisation cannot be aimed at a date. Pretending otherwise is how suppliers get into trouble.
A seasonal peak. Where demand exists for six weeks.
Building position for a term that matters briefly is poor value against buying it for the period it matters, unless the page is being built to last per our seasonal guide.
Testing whether a product sells. The most underrated use of advertising there is.
A week of spend answers a question that months of optimisation would answer expensively and slowly. Find out whether anybody buys it before investing in being found for it.
Clearing stock. Where the objective is moving units quickly rather than building anything.
Covering what optimisation has not reached. On a large catalogue there will always be areas the work has not got to.
Advertising covers them in the meantime, which is a sensible division rather than a compromise.
Where Optimisation Wins
Four cases where the balance runs the other way. All four are where a specialist shop makes its money.
Category terms. Somebody searching for a type of product rather than a specific item.
These carry the commercial search on a shop and they suit a category page far better than a product listing, since the shopper is still choosing. Our category page guide covers why.
Informational and comparison searches. How to choose, what the difference is, which one suits a situation.
Advertising serves these poorly because the searcher is not buying yet, so the click is expensive relative to its immediate value. A shop answering the question gets to be the one that supplies the answer. It costs nothing per visit.
The long tail across a large catalogue. The strongest structural argument here.
A catalogue produces an enormous number of low volume searches. Buying each of them individually is uneconomic. Being findable for all of them costs no more than being findable for some, which is a property advertising does not have.
Anything where the click price is high. Competitive categories where advertising is expensive.
The relationship is simple: the more a click costs, the more a free visit is worth. Per block four, an expensive category is exactly where a thin margin runs out of room first.
Running Both
The two work together better than either works alone, provided somebody is looking at them as one picture rather than two reports.
The division that works. Advertising covers what optimisation has not reached and what it cannot reach.
Specific products, seasonal peaks, new lines and anything time bound. Optimisation takes the categories, the informational searches and the tail.
What each teaches the other. More than most shops use.
Advertising data shows which terms convert before optimisation has reached them, which is the best prioritisation input available. Optimisation shows which terms you no longer need to buy, which reduces spend without reducing orders.
The double counting trap. The most common reporting error when both run.
A shopper clicks an advertisement, does not buy, returns two days later through a search result and buys. The advertising platform claims that sale. The analytics platform may also attribute it to organic traffic. Both reports are internally consistent and their totals exceed the shop's actual revenue.
How to catch it. Add the channels up and compare with the shop's own figure.
If the sum exceeds what the shop actually took, something is being counted twice. The excess tells you roughly how much. The shop's own order data is the only figure that cannot double count, which is why block nine treats it as the reference.
Comparing The Numbers Properly
Most comparisons between these two are unfair in ways nobody intended. Three rules make the comparison mean something.
Same window, both sides. The rule broken most often.
Advertising is usually judged over a month. Optimisation gets judged over the same month, in which it has produced almost nothing because it has been running for eight weeks. Compare both over a year. Compare over three. The picture inverts per block three.
Margin, not revenue. On both sides.
Revenue comparisons favour whichever channel sells the higher priced items, which is usually a fact about product mix rather than about the channel.
Fees included on both sides. The one that quietly flatters advertising.
Advertising cost is usually counted as the media spend alone, while optimisation is counted at the full fee. If somebody manages the advertising, that management cost belongs in the comparison. So does the time your own staff spend on it.
The reference. Your own order data, per block eight.
Not the advertising platform's figure and not the analytics figure. Both can be right about their own question and wrong about the total.
What a fair comparison usually shows. That the question was wrong.
The channels do different jobs at different stages. The shops that do best stop asking which is better.
Advertising scales.
This does not.
Twice the orders from advertising costs roughly twice as much. The same fee here produces considerably more in month twelve than in month four, because the work compounds rather than being consumed. A one year comparison favours advertising. A three year one usually does not.
What is included every month:
£350 per month, one target area. No setup fee, nothing billed separately.
Twenty-two guides.
One subject.
This guide covers the comparison with advertising. The rest of the series covers the sequence, structure, category and product pages, technical health, measurement and everything a store owner has to decide.
