The real ROI of SEO: how to calculate it for your business

Search for SEO ROI numbers and you'll find agencies advertising 700%+ returns, decade-old close-rate statistics passed around as current, and calculators that assume you'll rank #1 for everything. None of it is your ROI. SEO ROI isn't a number you look up — it's a short calculation you run with your own inputs, and this article is that calculation, start to finish.

Start from break-even, not from ROI

The textbook formula — profit attributable to SEO minus cost, divided by cost — is correct and useless as a starting point, because both inputs are foggy at the start. The useful opening question is smaller: how many new customers per month does SEO have to produce to pay for itself?

Two inputs: what you spend, and what a customer is worth. Say you spend $1,500 a month — mid-range for a real program, per the going rates — and an average customer brings you $800 in gross profit over their lifetime with you. Break-even is two customers a month. For a $200-profit customer it's eight a month; for a $5,000 client it's one every three months. That single number reframes the whole decision — you're no longer asking 'is SEO worth it', you're asking 'can content plausibly bring two buyers a month to my door', which is a question you can actually reason about.

Don't ask what SEO's ROI is. Ask how many customers a month pays for it. The second question has an answer you can check.

Use customer value, not first-invoice value

The most common way small businesses undercount SEO's return is valuing a customer at their first purchase. If your average customer returns, refers, or subscribes, the first invoice is the smallest number in the relationship. Compute lifetime gross profit honestly — average revenue per customer over however long they realistically stay, times your margin. This one correction routinely halves the break-even bar.

Forecasting: the math most people do dishonestly

Can content plausibly deliver those two customers a month? The forecast chain is traffic × conversion × your close rate, and every link invites optimism. Here's how to keep it honest.

• Traffic: click-through falls off a cliff by position. Backlinko's analysis of about four million search results puts position 1 at 27.6% of clicks, position 2 at 18.7%, position 3 at 10.2% — and page two at 0.63%, a rounding error. Forecast assuming positions 3–10 on your keywords, not #1, and assume a share of your informational queries lose clicks to AI answers.
• Conversion: the share of visitors who take your money action — a call, a quote request, a signup. Benchmarks range from roughly 2% to 8% depending on industry and how you measure, which is too wide to borrow. Assume 1–2% until your own data says otherwise; late-stage pages convert multiples better than informational posts, which is why knowing which posts sell changes the math.
• Close rate: yours, from your own sales history. Not an industry stat — more on that below.

Worked through: suppose twenty ranking articles eventually draw 2,000 organic visits a month, 1.5% request a quote, and you close a third. That's ten leads and three customers a month — comfortably past a two-customer break-even, with room to be half wrong. If the same math needs everything to go right to reach break-even, that's your answer too, and it's cheaper to learn it on paper.

Count the real costs

The denominator is more than the invoice. Money: the retainer, software, or per-article fees — whichever pricing model you're under. Time: your hours reviewing, briefing, and editing, priced at what your hour is actually worth. A $99 software subscription that consumes ten founder-hours a month is not a $99 channel. Understating cost is the polite way businesses lie to themselves about ROI — and the favorite way vendors lie to them.

The timing problem: why early ROI always looks terrible

SEO front-loads cost and back-loads return. Months one to three produce spend and little else; measured at month three, almost every SEO investment shows negative ROI, and measured at month eighteen, successful ones look absurdly good — because articles keep producing after you've stopped paying for them, which is the property paid ads never have. So judge the investment on a payback horizon, not a monthly one: expect six to twelve months to break even on a program that's working, and use the early signals — impressions, query growth, position drift — to tell a slow success from a failure long before the revenue math resolves.

Measure it for real once it's running

Forecasts start the decision; measurement finishes it. That means conversions connected to the organic channel, tracked per landing page, reviewed quarterly. Once that plumbing exists, your ROI calculation stops using assumed conversion rates and starts using yours — and the forecast above becomes a quarterly report instead of a bet.

Three ROI numbers to ignore

• The 14.6% close rate. You'll see 'SEO leads close at 14.6% vs 1.7% for outbound' everywhere. It traces to HubSpot research from around 2011 — recycled for fifteen years without revalidation. Whatever your close rate is, it isn't a number from 2011.
• Agency ROI averages. '700% ROI over three years' comes from agencies measuring their own client campaigns — the successful, multi-year ones. It's a portfolio highlight reel, not your expected value.
• Fantasy calculators. Any projection multiplying total search volume by #1-position click-through by an optimistic conversion rate produces a large number and no information. If a proposal's ROI projection assumes top rankings across the board, the projection is the first deliverable they've failed.

The pattern: every borrowed ROI number serves the person quoting it. The calculation in this article runs on numbers you own — your spend, your customer value, your close rate — and that's precisely what makes it trustworthy enough to act on.