What a Good ROI for SEO Looks Like, and How to Read Yours.

A good ROI for SEO usually lands somewhere between 300% and 750% once a campaign has had a year or two to mature, which means every rupiah you put in comes back three to seven times over. A handful of industries do far better than that, and a badly run program can sit near zero for a long time before it turns. So the honest answer to what counts as good depends on your industry, how patient your business can afford to be, and how you decide to count the revenue in the first place. In this post I want to walk through the benchmarks I use with enterprise clients, how long the payback takes, what SEO costs in the Indonesian market, and why the old way of counting SEO returns is starting to break as more buying decisions happen inside AI answers before anyone clicks anything.

The math behind SEO ROI

The formula itself is not complicated. You take the revenue that organic search brought in, subtract what you spent to earn it, divide that by the same cost, and multiply by a hundred. Spend Rp 100 million over a year and attribute Rp 500 million in revenue to organic, and your ROI is 400%. The arithmetic is the easy part, and it is never where teams get stuck.

The hard part is deciding which revenue counts as organic, over what window, and how much credit search deserves when a customer touched five other things before they bought. A buyer might find you through a Google search in March, follow you on LinkedIn, come back through a branded search in June, and convert after a sales call in August. A strict last-click model hands all of that revenue to the sales call and tells you SEO earned nothing. That is why SEO ROI looks lumpy and unconvincing in the first few months and then compounds hard later, because the content and authority you built keep pulling in demand long after you paid for them.

What good looks like by industry

The most useful public benchmark I point clients to comes from First Page Sage, which tracks SEO returns across industries over several years. Their 2026 report puts a well-run campaign at roughly a 748% average return, and the spread by sector is wide enough that a single industry number tells you more than the overall average does.

The pattern under those numbers is straightforward once you see it. Industries with high deal values and long customer lifetimes post the biggest returns because a single won client covers many months of SEO cost, so real estate and financial services sit at the top. eCommerce runs lower per transaction, but the volume and repeat purchases still make the case comfortably. What you are looking at is a floor of "a few times your money back" and a ceiling that, in the right sector, reaches ten or fifteen times.

Service type moves the number just as much as industry. First Page Sage separates technical SEO on its own, which returns around 117%, from content and thought-leadership work, which is where that 748% figure comes from. Technical fixes stop your site from leaking, and that matters, but the compounding returns come from owning the topics your buyers search, which is the whole reason we built our Topic Ownership Strategy around clusters rather than one-off pages.

Why SEO tends to beat paid over time

Organic search still drives more of the web than any other channel, and it is not close. A widely cited BrightEdge study found organic search responsible for 53% of all site traffic, against 15% for paid, and organic leads tend to convert better once they land. First Page Sage reports that in financial services, SEO converts more than seven times better than paid search. The difference in returns comes down to what you are paying for, because paid search rents attention and stops the moment you stop paying, while an SEO asset you build this quarter keeps bringing in traffic and leads for years without a per-click fee attached to each visit. If you need qualified leads next week, paid is the right tool and I will tell a client so. Over a two or three year horizon SEO wins on ROI because you are buying an asset instead of renting clicks, and the asset does not send you an invoice every time someone finds it.

How long before it pays for itself

Most campaigns reach positive ROI somewhere in a 6 to 12 month window, and the exact break-even depends on the mix of work. Technical cleanups can pay back inside six or seven months, while heavier content programs often take closer to a year before the revenue catches the spend. Peak returns usually arrive in the second or third year, once the authority has accumulated and the content library is broad enough to rank for a lot of related searches at once.

This is the part that decides whether an SEO investment survives inside a company, and it has less to do with the tactics than with expectations. If your finance team is told to expect returns in month three, a program that is performing perfectly normally will look like a failure in month four and get cut right before it compounds. Being honest about the timeline from the start is what keeps a good campaign alive long enough to pay.

What SEO costs in the Indonesian market

Numbers only mean something once you can put your own cost into them, so these are the rates SEO tends to run at in Indonesia. Entry-level and local work sits around Rp 2 million to Rp 8 million a month. A growth-stage program lands roughly between Rp 8 million and Rp 20 million a month. Enterprise campaigns run from Rp 20 million up to about Rp 150 million a month, and at the top of that range one strategy team is usually coordinating several brands or markets so they share research, tooling, and earned authority instead of paying to build the same foundation three times over.

Put a real figure through the formula and the case gets concrete. Say a growth-stage retailer spends Rp 15 million a month, which is Rp 180 million across the year. At a conservative 3x return, the kind eCommerce posts on the low end, that is Rp 540 million in attributed revenue, for a net gain of Rp 360 million. In a higher-margin services business closer to the financial-services benchmark, the same spend is defending a far larger number.

The mistake I see most often is a business anchoring on the monthly fee instead of the ratio. Rp 20 million a month feels expensive in isolation, and it stops feeling expensive the moment you hold it against the revenue a fully ranked topic cluster brings in over a year.

Why some campaigns never reach these numbers

The averages hide a wide floor, and the programs that sit near the bottom of it rarely fail for technical reasons. The most common problem is chasing traffic that was never going to buy, ranking a business for broad high-volume terms that pull in visitors with no intent to purchase, which inflates the traffic chart and does nothing for revenue. A good return starts with targeting the searches your actual buyers make, even when the volume looks small next to a vanity keyword.

The second is a page that ranks but does not sell. Plenty of sites win the position and then hand the visitor a page with no clear next step, no reason to trust the brand, and no route to a quote or a cart, so the traffic arrives and leaves. Search can only deliver the visit, and the return depends on what the page does with it once the visitor is there.

The third is measuring the wrong thing entirely. When a team reports rankings and sessions instead of leads and revenue, a campaign can look busy for a full year while returning almost nothing, and nobody catches it because the dashboard was never pointed at money in the first place.

Why the old ROI math is starting to break

Search is increasingly answering the question on the results page itself, so the click that every SEO ROI model was built to count is disappearing, and that changes how all of this has to be measured. Studies now put zero-click searches near 68% in early 2026, and a controlled field study from researchers at the Indian School of Business and Carnegie Mellon found that Google AI Overviews cut organic clicks by 38% on the queries where they appeared, with those overviews showing up on more than 40% of searches.

If a chunk of the value your SEO creates now shows up as a mention or a citation inside an AI answer that never sends a click, a last-click ROI model will under-count it and tell you your returns are falling even while your real influence is climbing. The work that gets your brand named and cited by ChatGPT, Gemini, and Google's AI Overviews is the same SEO work that used to earn the click, so the return did not vanish, it just stopped showing up in the one report most teams still rely on.

How to measure return when the click disappears

The fix is to count what the AI answer does with your brand rather than only the traffic it forwards. We track four things for clients, which are how often your brand gets mentioned across the major assistants, your share of the citations against named competitors, the sentiment of how you are described, and the AI-attributed traffic that does still arrive. Together they show whether you are winning the answer even on the searches that never produce a click, and we publish the full measurement methodology openly rather than hiding it behind a proprietary black box.

The returns from measuring this way can be large because the space is still uncontested. On one serviced-residence campaign we took a hospitality brand's AI-answer audience from 7.4 million to 34.6 million in a single quarter, a jump of roughly 450%, by making the brand the most cited option for the questions its buyers were asking assistants. That kind of movement never appears in a clicks-only report, which is exactly why the brands that measure it early are compounding a lead while their competitors are still arguing about whether SEO is worth the line item.

If you want to know what your own SEO is returning in an AI-first search world, that is the work our AI visibility measurement is built to answer, starting with a clear read on where your revenue is coming from before we touch a single page.