When I started my agency back in 2023, my first prospect was a B2B SaaS company. I was so excited to land the first SEO client for my consultancy. It didn’t work out, and I ended up turning the client away.
The company was brand new and preparing for its first round of funding. As part of the proposal process, I asked questions to understand the product and define the audience. The product was ahead of its time: it solved a problem that didn’t really exist yet, and there was little to no search volume for keywords related to it.
Now, that alone isn’t a reason to say no. In fact, it can be a strong reason to say yes to SEO. SEO coupled with content marketing can create demand for products and services. Sometimes clients don’t know they need something until they read about it.
I had a strong strategy in place. But the client needed a return in three months. And although that is possible in many scenarios with SEO, it wasn’t possible for this client. So I had to decline and tell him that SEO wasn’t the route to a quick return in his situation. Instead, I pointed him toward partnering with B2B sites that could get his foot in the door through the partner, rather than through search.
Here’s the bottom line: SEO is worth it when a customer is worth enough, when people are actually searching for what you sell, and when you can wait six to twelve months for the compounding to start. If one of those three isn’t true, it usually isn’t worth it yet, and we’d rather tell you where the money should go instead.
The short answer
SEO tends to pay when three conditions line up:
- A customer is worth enough. High lifetime value, or repeat business, means a small number of organic leads covers the fee. This matters most because SEO costs roughly the same whether you sell a $200 service or a $20,000 one.
- Search demand exists. People already type your problem into Google. SEO captures demand; it doesn’t create it.
- You can wait. Organic results compound, which means the early months are the expensive ones and the later months are the cheap ones. If you need the payback in a quarter, the math rarely closes.
And it tends not to pay when:
- You’re still figuring out the offer. Rankings built around a service you drop in six months are sunk cost.
- Nobody can implement. Recommendations that sit in a PDF produce nothing.
- You need leads this month. SEO is the wrong tool for a cash crunch. Paid search or outbound is.
The break-even math
Four numbers, no spreadsheet:
Customers needed = (monthly fee × months) ÷ gross profit per customer
Gross profit, not revenue: what’s left from a sale after the direct cost of delivering it, because that’s the only money that can actually pay an SEO fee. Then the question that matters: can organic search plausibly send you that many customers in that window?
Here’s one worked example. The inputs are illustrative; swap in your own.
| Input | Illustrative value |
|---|---|
| Monthly SEO fee | $2,500 |
| Evaluation window | 12 months |
| Total spend | $30,000 |
| Gross profit per new customer | $1,500 |
| Customers needed to break even | 20 (about 1.7 a month) |
| Your close rate from qualified leads | 30% |
| Organic leads needed to break even | about 67 (roughly 5 to 6 a month) |
Five or six qualified organic leads a month, by the end of year one, for a business where each closed customer is worth $1,500 in gross profit. For a commercial HVAC contractor or a specialty law practice, that’s a modest target. For a business where a customer is worth $150, the same math needs 200 customers, and now we’re asking whether your market even contains that many searchers.
Two caveats. Leads don’t arrive evenly, they ramp, so the first months contribute far less than the last. And a customer acquired through search often comes back, which the model ignores in your favour.
When SEO is not worth it
We’d rather say this plainly than have you find out at month nine.
Before product-market fit. SEO builds an asset around specific pages and specific search terms. Change the offer and the asset is stranded. We ask about the roadmap on a first call because a pivot in the plan is a reason to wait.
Low customer value with no repeat business. If gross profit per customer is under a few hundred dollars and they don’t come back, the break-even count gets so high that search volume itself becomes the ceiling. Not impossible, but a harder case, and one where a paid channel with cheaper clicks may fit better.
Leads needed within 30 days. Google’s own guidance on hiring an SEO, “Do you need an SEO?” (updated June 2026), is blunt that no one can guarantee a #1 ranking, and the timing is no more controllable than the position. Paid search can be live tomorrow. SEO can’t.
Nobody to implement changes. Most of an engagement’s value is in changes to the site: new pages, rewritten pages, technical fixes. If there’s no developer, no writer and no owner with an afternoon a week, the recommendations don’t ship, and unshipped recommendations have an ROI of zero. This is the most common reason we see SEO that “isn’t working”.
No search demand. A truly new category has no queries to rank for yet. Build demand with other channels first, then come back to capture it.
What the ROI studies say, and their limits
The number you’ll see quoted everywhere is First Page Sage’s 748% average ROI over three years. It’s worth knowing where it comes from: that agency’s own client campaigns, run between Q1 2021 and Q3 2025. It’s a real figure. But it describes funded, multi-year campaigns that survived long enough to be measured.
That’s the limit. A campaign that got cancelled at month four because the owner ran out of patience isn’t in the average. Neither is the one where the client never implemented the recommendations. Averages drawn from surviving campaigns describe the upside for a business that commits for years and executes, not the expected outcome for someone deciding whether to start. Read 748% as “this is what the top of the distribution looks like,” not as a forecast.
Run your break-even math, and treat any ROI study as a sanity check that the upside exists, which it clearly does.
How long it takes to pay back
Ahrefs’ May 2025 study of how long it takes to rank found that only 1.74% of new pages reach Google’s top 10 within a year, and that 72.9% of the pages sitting in the top 10 are more than three years old. Those numbers sound bleak until you read the next one: of the new pages that did reach the top 10, 40.82% got there within their first month.
All that to say, most new pages don’t rank, but a well-targeted page on a site with some authority can move fast. The spread is enormous, and which side of it you land on depends on the site you start with and the terms you pick. That’s why the “six to twelve months” answer is honest rather than evasive: the early lifts often come in weeks, the break-even usually doesn’t.
Across our own engagements, here’s the honest version. Of the 11 engagements since 2024 that we could measure from a clean baseline, 8 reached a sustained lift of 20% or more in estimated organic traffic, and 3 never did, including two that ended early. At month six, the median engagement was up about a third on its pre-start baseline (+33%, across 10 engagements), with a range from a 21% drop to nearly triple. We can’t give you a break-even month from our own data yet, because we don’t have revenue figures for most clients. That’s the number we’re now working to track, and we’ll add it to our research when we have it.
Methodology (TFS engagements). We took every TFS engagement from January 2024 to September 2026 with six or more months of fees (27) and kept the 11 where we could measure the whole site from before we started. The metric is Semrush’s estimated organic traffic, not Search Console clicks, so it includes branded searches and can’t separate our work from market trends or Google updates. “Lift” means the three-month average ran at least 20% above the three months before we started and stayed above that baseline for two more months. Work on one section of a large site, white-label work, and new sites with no baseline are excluded. With 11 engagements, one account moves the median.
Has AI search changed the answer?
Partly, and in a direction that makes revenue-driven SEO matter more, not less.
Ahrefs’ analysis found that AI Overviews are associated with a 58% lower click-through rate for the top-ranking organic page. That’s an association, not a law, but it matches what we see: when Google answers a question on the results page, fewer people click through to the page that would have answered it.
The mechanism matters for your decision. The queries AI Overviews answer well are informational ones, top-of-funnel: “how does a heat pump work,” “what is a living trust.” Those posts were the weakest link to revenue to begin with. The queries that hold up are the ones with a transaction or a location attached: “heat pump installation near me,” “estate planning attorney in Tempe.” Someone searching those wants a business, not a paragraph, and an AI summary doesn’t install the heat pump.
So if your past SEO spend went into a blog that ranked for questions, the ROI on that has probably fallen. If it goes into your money pages, the service pages where a click can become a call, we’d guess the effect is much smaller. We’re not going to predict where AI search lands next year. We can say that the parts of SEO tied directly to revenue are the parts we’re most confident still pay.
How to measure it
Sessions and rankings are not results. Measure SEO the way you’d measure a salesperson: what did it close?
Tracked conversions from organic. Form submissions and quote requests, attributed to organic search in your analytics, so you can see the source of each lead.
Call tracking. For most local and service businesses, the phone is the money channel. A tracking number swapped in for organic visitors tells you which calls search sent.
The CRM. Leads go in with their source; the CRM tells you which ones became booked jobs, and what they were worth. That’s the number that goes back into the break-even formula.
The report you want each month is booked jobs or pipeline from organic, against the fee. If the person you’re paying only sends traffic charts, ask for a different report.
What we see in practice
For the client I mentioned at the start, the math didn’t work:
- $20 a month per user, with $16 a month in profit
- An SEO fee of $2,500 a month
SEO would need to bring in about 160 new users every month just to break even. That isn’t impossible, except the search demand for this particular product was extremely low. I could get them to rank, but I wouldn’t be able to bring in 160 new users every month.
If you want us to run these numbers for your business, book a free consultation. If the answer is no, we’ll say so.
Questions, answered
Is SEO worth it for a small business?
Often, and sometimes not. It comes down to what a customer is worth, whether people search for what you sell, and whether you can wait. Google’s own guidance notes a small local business can do much of the basic work itself, so for very small operations, self-serve first and paid help once the math supports it.
Is SEO dead?
No, but part of it has weakened. Ahrefs’ analysis links AI Overviews to a 58% lower click-through rate on the top organic result, which mostly hits informational queries. Searches with buying or local intent still send people to businesses. SEO aimed at those money pages is alive; SEO aimed only at ranking for questions is worth less than it was.
Is SEO better value than paid search?
Different tools. Paid search is faster and stops the moment you stop paying; SEO is slower and keeps working after the fee is spent. Early on, paid usually costs less per lead. Over a couple of years, organic often wins, provided it targets revenue terms. Many businesses run both and shift budget as organic grows.
What is a good ROI for SEO?
Any return that clears your break-even within the window you can afford. First Page Sage’s 748% three-year figure is drawn from their own surviving client campaigns, so treat it as the upside, not the expectation. A more useful target: cover the fee in gross profit from organic customers by month twelve, then compound from there.
How soon will I know if it’s working?
Leading indicators show up first: pages indexed, impressions rising, rankings moving on the terms you chose. Those can appear within weeks. Leads follow, usually within a few months, and break-even comes later. If nothing has moved on impressions or rankings by month four or five, something is wrong with the targeting or the implementation, and it’s worth asking why.
