Why My SEO ROI Is Negative

Attribution windows were built for a media environment where a click and a purchase happen close together. Organic search rarely works that way. A visitor researching a purchase might read several pieces of content over weeks before ever filling out a form, and the deal itself can take months more to close after that first conversion. Most attribution tools ship with a default window sized for paid social or paid search, and teams keep that default because nobody had a reason to question it.

When the window closes before that cycle finishes, every deal that takes longer gets counted as a loss on the SEO side of the ledger, even though the channel did the work that started it. Finance sees a channel with rising cost and flat credited revenue, and the natural conclusion is that the channel stopped working. The fix is not to trust SEO more. It is to measure the actual days-to-close for organic-sourced deals and size the window to match, instead of reusing a window built for a faster channel.

Blended customer acquisition cost adds up every dollar of marketing spend and divides it by every closed customer, regardless of which channel actually produced them. It is a fast number to calculate and a poor one to judge a single channel by, since it assumes every dollar spent contributed equally to every customer won. Finance teams like it because it is simple, but simple and accurate are different properties.

A quarter with a large brand campaign or a paid search push raises the blended number for everyone, including customers who found you through organic search and never touched a paid channel. SEO ends up carrying cost it did not create, and the resulting ROI figure penalizes content work for a media decision it had no part in. Separating incremental, channel-specific cost from shared or unrelated spend is what turns a blended figure into one that actually answers whether SEO on its own is profitable.

Most finance and marketing reporting books cost in the month it is spent. A content sprint that costs real money to research, write, and publish shows up as a lump expense in that single month, then gets compared against whatever revenue closed in that same period. Content almost never works on that clock, and treating it as if it does sets the comparison up to fail before any writer touches a draft.

A page published this month may keep earning organic visits and contributing to closed deals for a year or more after that first publish date. Judging it against one month of revenue comes close to guaranteeing a negative result, no matter how well it eventually performs over its full working life. Amortizing production cost across the asset’s realistic working life, and tracking cumulative revenue against that spread cost, is the version of the calculation that actually answers whether the content paid for itself.

Traffic metrics are easy to pull and easy to report, so they end up standing in for a payback calculation that never actually happens. Cost per click and cost per session describe how expensive attention was. They say nothing about how long it takes the resulting customer’s gross margin to repay what was spent acquiring them, which is the number a unit-economics review actually needs.

Payback period is the number that decides whether a channel is a good investment, not the number that decides whether it drove volume. A channel with a higher cost per click and a short payback can be a far better investment than a cheaper channel with a long one, even though the cheaper channel looks better on a cost-per-click slide. Without that calculation, ROI conversations default to whichever traffic metric is loudest, and volume gets treated as if it were profitability instead of a precondition for it.

SEO’s job, in most unit-economics conversations, is acquiring new customers who would not otherwise have found you. Expansion revenue from existing accounts is real revenue, but it usually comes from account management, product usage, or renewal motions that have little to do with organic content, and crediting it to SEO overstates what the channel is actually contributing.

When both revenue types land in one line, a good quarter for expansion can mask a genuinely weak quarter for new-logo acquisition, and a strong content quarter can get no credit because expansion happened to dip at the same time. Leadership sees one flat or declining number and assumes SEO stalled, when the acquisition motion it drives may have been fine the whole time. Isolating new-logo revenue by source is what lets SEO’s actual acquisition contribution be judged on its own terms instead of averaged against a motion it does not drive and was never meant to influence.

Ranking for a high-volume keyword feels like a win, and the session count backs that up. Volume alone says nothing about the buying stage or fit of the visitors that keyword attracts. A broad, top-of-funnel term can bring in browsers, students, and low-intent researchers alongside genuine buyers, and all of them count equally in a traffic report that only measures sessions and rank position.

If a meaningful share of the resulting customers churn quickly or land on the smallest available plan, their lifetime value may never clear what it cost to acquire them, even at a low cost per click. A channel can look cheap and still be unprofitable once the customers it produced stop paying. The unit-economics fix is not to abandon the keyword. It is to segment lifetime value by the landing page and query that sourced the customer, so ROI gets judged on the customers actually produced instead of the traffic volume alone.