SEO pays off when somebody genuinely searches for what you sell, one customer is worth real money to you and you can wait a few months. The standard answer here is “it depends” and that is usually where it ends - which is the worst possible answer. Let me disarm it straight away: it depends on exactly those three things, and you can check each one in an hour, before you spend a single zloty.
This piece is a summary of a conversation I have with clients roughly once a week. You will not find promises or graphs pointing up - you will find a way of doing the maths that you can apply to your own business, and an honest list of situations in which I talk people out of SEO and AI visibility work myself.
Three questions every calculation starts with
Before you count anything, answer three questions. If the answer to any of them is “no”, counting further is pointless.
- Does anybody actually type what you sell into a search box? Not “do people need my product”, but do they call it what you call it. You can check that in a quarter of an hour: type into Google what you think a customer would type, and see whether competitors come up or random content.
- What is one customer worth? Not the value of a single invoice, but of everything that customer leaves with you across the whole relationship. A one-off purchase and a customer coming back every month for three years are entirely different sums.
- How long can you wait? Honestly, looking at your bank account - three months, a year, or “I need orders this month”? That question settles more than the previous two together.
The maths in plain language - how to run it for yourself
I will not hand you ready figures, because every industry counts differently and invented numbers do more harm than good. I will give you the method - the examples below are openly notional, so drop your own data into them.
Step one: work out the value of one customer. Say an average job leaves you X in profit after costs, and the average customer comes back twice. The customer value is 2X. With a subscription service you multiply the monthly amount by the realistic number of months the relationship lasts - not the dream number.
Step two: decide how many customers a month would change your situation. That is a more important question than “how many visits do I want on the site”. Sometimes the answer is “two”, and then things look completely different from fifty orders a day. Multiply that number by the customer value and you have the monthly amount SEO would have to deliver.
Step three: compare it with the annual cost of visibility work. Annual, not monthly, because SEO proves nothing over a quarter. Into the sum go technical work (speed, indexing, structure), writing content for twelve months and fixing whatever is blocking the site at the start. There are three billing models: a monthly retainer (predictable, but you have to know what it covers), a project with a closed scope (good for foundations and content), and payment on results. That last one sounds safest and is the trickiest: it is easy to bill against phrases nobody types but that look nice in a report.
If the amount from step two is clearly larger than the annual cost from step three, the maths works. If it comes out level, SEO will be an expensive hobby.
What helps: before you order anything, go into Google Search Console and check which queries your site already shows up for. It is free and it tells you more than any presentation: if you can see buying-intent phrases from your industry there, you have a foundation. If all you see is your own company name, you first have to build a reason for anyone to find you.
When SEO does not pay off - said plainly
There are situations where I advise against SEO even though it costs me the job. The four most common:
- Nobody is searching for it. A new product category you have to name yourself first, or a niche thing sold through relationships and referrals. A search engine answers existing demand - it does not create it.
- You need customers this month. SEO is a terrible rescue tool. If the orders have to be there immediately, the money should go to something that switches on straight away.
- The site is in a state where every zloty leaks out. If the cart falls apart, the form does not send emails, the phone number is hidden in the footer and the site takes a dozen seconds to load - driving traffic to it is pouring water into a leaking bucket. Fix first, visibility second. How to check whether Google can see the site at all I described in the piece on why your site is not showing up in Google.
- The budget only covers three months. The saddest case, because the money goes on sensible work - the result simply arrives after the budget has run out. That is paying for the foundations and leaving before the roof. Better to put it aside and do it properly in six months.
When SEO does pay off
The inverse of that list is just as concrete. SEO usually makes sense when:
- Demand is repeatable and expressed in words. People regularly type the name of what you do into a search box - faults, services, specific products.
- One customer is worth a lot. Then a handful of jobs a year from search is enough to pay the investment back. Which is why SEO works so well in B2B services and on higher-value projects.
- You operate locally. Queries with a city name usually face less competition than nationwide ones, and the buying intent in them tends to be clearer.
- Content works for years. A good piece answering a customer’s question brings people in long after you paid for it. It is one of the few marketing channels that behaves like an asset rather than a cost.
- You have a store with a specific range. Categories and products people search for by name are natural entry points from search - I wrote about when a store’s maths works in the piece on whether an online store is worth it.
How long you have to wait for results
Months, not weeks. That is not an excuse, it is how a search engine works: it has to notice the changes, process the new content and check whether people are happy with it.
What matters more is what you can see along the way. First impressions grow - the site shows up for more queries, even if still a long way down. Then new phrases appear that you were not present for at all. Only at the end come the clicks and the enquiries. If after a few months you cannot see even the first signal, that is not a matter of patience, it is a sign that something is wrong with the work or with the assumptions. How to track it yourself I show in the piece on how to check your Google rankings.
SEO versus paid ads - this is not a religion
The difference is simple. Advertising buys traffic immediately and ends exactly when you stop paying - it is a tap, not a well. SEO works the other way round: for the first few months you pay and instead of sales you get signals - impressions and new phrases - and only then do you have an asset that keeps working even when you ease off.
Which is why the combination usually makes sense: ads keep sales going here and now and quickly show which phrases genuinely bring customers, while SEO builds in the background the channel that will let you come off the ad budget a year from now. Treating either one as the only true path costs more than either choice.
New in the calculation: part of the traffic is moving to AI
More and more people do not browse a list of results - they ask ChatGPT or read the AI-generated answer above Google’s results. That traffic still comes from the same content, because AI assembles its answer from specific pages, it is simply harder to account for. Search Console now reports impressions from AI-generated answers separately, but without clicks and without queries, so the classic “position for a phrase” stops being enough as a measure.
The conclusion is that the calculation has to be broader: not “am I first for phrase X”, but “am I present where people look for answers”. I set that out separately in the piece on SEO and visibility in AI - the foundation is the same, so the work does not go to waste.
How not to burn money
Four rules I repeat to everyone, regardless of who ends up doing the work:
- Start with the foundations. Speed, correct indexing, an orderly description of the site for machines, and content that answers real questions. It is boring and makes no impression in a presentation, but without it the rest has nothing to stand on. Where to start the clean-up I set out in the piece on what an SEO audit covers, and the technical description layer in the piece on structured data.
- Measure in Search Console. A free source of truth about what you show up for and what is changing. A supplier’s report is an interpretation; Search Console is a fact.
- Do not buy a package with no defined scope. “SEO for this much a month” is not a scope. A scope is a list of things that will be done in a given month, and the way you will check them.
- Run from guarantees of first place. Nobody controls search results - not me, not anyone. That promise means either naivety or a phrase so narrow that nobody types it.
If you do not know where you stand
The most common reason for a burned SEO budget is not a bad supplier but the absence of a starting point - a decision made without knowing whether the site is fit for visibility and whether the demand exists.
So before you order anything, start with a free diagnosis: you send me a link, I come back with specifics - whether the site has a foundation, where you stand in search and whether the maths has a chance of working. I will say plainly when the answer is “in your position I would not go into this” - it costs me the job, but it saves you a year of paying for something that was never going to work.
Frequently asked questions
Is SEO worth it for a small company? Often yes, because a small company targets narrow and local queries where the competition is far smaller. There is one condition: work out the value of one customer and check whether a few customers a year from search cover the annual cost of the work.
How long before SEO results show? Sales results are counted in months, not weeks. Before that you see indirect signals - growing impressions and new phrases in Search Console - and that is what you use to check whether the work is going the right way.
Which is better: SEO or Google Ads? It is not an either-or. Advertising gives traffic immediately and stops with the budget; SEO builds an asset that stays. It usually makes sense to run ads for now and build organic visibility for later in parallel.
Can first place in Google be guaranteed? No - nobody controls search results. Influencing positions, yes; but a guarantee of a specific place is a promise that cannot be kept, and it usually concerns a phrase so narrow nobody types it. An honest offer talks about the scope of work, not a promised position.
Want your company to show up in Google and in ChatGPT answers? I work on technical SEO and AI visibility from the code side, not the reporting side. Tell me which questions matter to you and I will quote it.