A campaign at 149 USD a month is either trivially cheap or impossible to justify, depending entirely on what you divide it by. This article works through both tiers, both add-ons and one fully specified twelve-month example — and then argues that for a Finnish exporter the only sensible denominator is the contract, not the click.
The usual case for a search campaign is built on volume. So many searches a month, such a click-through rate at such a position, therefore so many visits, therefore a cost per visit you can compare against paid channels. It is a clean argument and it works for a retailer.
It collapses for a company in Helsinki selling ice-class propulsion refits, web-tension control systems or calibration equipment to eleven countries. The volume side of that equation is close to zero. The value side is enormous. A single order can carry six figures, and the year's entire search demand for the term that produced it might be four hundred queries worldwide.
The difference is not features, it is who decides
Both levels of My SEO run the same machinery underneath: a keyword pool fed from Search Console, live result pages and your own seed terms; automated link building through the partner network; on-site suggestions; the full analytics stack; and the live assistant. Nothing in that list is withheld from the cheaper tier.
What the higher tier buys is authority over the automation and people to exercise it. Keywords can be selected by hand, with the automatic pool as a fallback rather than the only source. Link placement can be directed at donors above a chosen domain-rating threshold instead of accepted as the system finds them. On-site changes can be routed through human review before they go live.
| Capability | AutoSEO | FullSEO |
|---|---|---|
| Keyword discovery and prioritisation | Automatic | Manual selection, automatic fallback |
| Backlink building | Automatic placement | Manual placement with a DR target |
| On-site suggestions | Applied from AI proposals | Human-review mode before publication |
| Search Console and result-page analytics | Full | Full |
| Live assistant in the campaign feed | Included | Included |
| Specialists, developers and writers | Not included | Included |
That last row is the one that decides most upgrades, and it is easy to skim past. The higher tier is not only software with more switches exposed; it carries a team of SEO specialists, developers and writers alongside the automation. For a twelve-person engineering firm with no marketing department, that is usually the actual purchase.
What the two tiers cost on their own
Pricing is per domain per month, which matters immediately for anyone running separate country sites or a legacy domain alongside the main one. Two domains means two subscriptions; there is no portfolio rate that changes the arithmetic below.
Held flat for a year and with nothing added, AutoSEO comes to 1,788 USD and FullSEO to 6,000 USD. The gap between them is 351 USD a month, or 4,212 USD across twelve months — roughly the cost of two days of a specialist contractor in Finland, which is a useful thing to hold in mind when the upgrade is being argued about.
Semalt sets out how the two campaign levels are structured in more detail. The figures above are the whole of the base price; everything that follows is optional and priced separately.
Two add-ons, both priced per slot
A slot is one placement. Both add-ons are sold in fixed steps rather than as a free quantity, and the two differ by a factor of ten in unit price for reasons worth understanding before you buy either.
| Add-on | Price per slot | Available steps | Monthly cost at each step |
|---|---|---|---|
| Wikipedia placements | 10 USD | 0 / 1 / 5 / 10 | 0 / 10 / 50 / 100 USD |
| PBN placements | 1 USD | 0 / 20 / 100 / 500 | 0 / 20 / 100 / 500 USD |
The arithmetic is straightforward in both rows: ten slots of the first at 10 USD each is 100 USD a month, and five hundred slots of the second at 1 USD each is 500 USD a month. Taken together with the higher tier, the maximum configuration for a single domain is 500 plus 100 plus 500, which is 1,100 USD a month, or 13,200 USD across a year.
The broader network behind all of this runs to more than 230,000 partner sites, which is what makes the automatic tier viable at its price. Where the donors come from and what the reporting shows is covered in the description of automatic placement.
The two tiers side by side
AutoSEO — the campaign runs itself
For a company with no marketing capacity that wants continuous activity rather than supervision.
- Keywords are found and ranked for you. Candidates arrive from Search Console, live result pages and your own seed terms, and the system decides the order of attack.
- Links are placed without a queue to approve. Placement proceeds continuously through the partner network, which is the main reason the tier can be priced where it is.
- On-site proposals arrive as suggestions. Titles, headings and internal linking are proposed by the model against your actual pages rather than against a generic checklist.
- Analytics are not reduced. The Search Console views, the result-page tracking and the assistant are the same ones the higher tier uses.
FullSEO — automation with a hand on it
For companies whose terminology is technical enough that an automatic keyword pool will get it wrong.
- Every candidate is approved, rejected or deferred. The pool still fills automatically, but each term passes through a decision, which is what stops a campaign optimising a phrase nobody in your industry uses.
- Placement can be aimed. A domain-rating threshold is set and links are placed manually against it, rather than taken as the network offers them.
- Changes wait for a person. Human-review mode holds on-site edits until someone signs them off, which matters where a specification sheet is a legal document.
- People come with it. Specialists, developers and writers work alongside the automation, which is the part a small engineering firm cannot replicate internally.
The approval queue is the feature that repays attention on a technical site. Finnish compounds and case endings scatter one concept across many strings, and an automatic pool will happily prioritise a grammatically plausible phrase that no procurement engineer has ever typed. Working through the manual keyword queue once a fortnight is cheap insurance against that.
A twelve-month example, costed line by line
What follows is a single constructed scenario for one domain: a Helsinki manufacturer of measurement and testing equipment, exporting to about a dozen countries. Three months on the automatic tier to see whether anything moves, then nine months on the full tier with add-ons.
| Phase | Months | Tier | Wikipedia | PBN | Per month | Phase total |
|---|---|---|---|---|---|---|
| Calibration | 1–3 | AutoSEO, 149 | 0 slots, 0 | 20 slots, 20 | 169 USD | 507 USD |
| Escalation | 4–12 | FullSEO, 500 | 5 slots, 50 | 100 slots, 100 | 650 USD | 5,850 USD |
| Full year | 1–12 | — | — | — | 529.75 avg | 6,357 USD |
Check the two phase totals directly. 169 multiplied by three months is 507. 650 multiplied by nine months is 5,850. Added together, 507 plus 5,850 is 6,357 USD for the year, which divided by twelve is an average of 529.75 USD a month.
The same figure should also be reachable by line item, and it is worth doing once because a mismatch here usually means a slot step was misread.
| Line | Monthly rate | Months | Subtotal |
|---|---|---|---|
| AutoSEO subscription | 149 USD | 3 | 447 USD |
| FullSEO subscription | 500 USD | 9 | 4,500 USD |
| Wikipedia, 5 slots | 50 USD | 9 | 450 USD |
| PBN, 20 slots | 20 USD | 3 | 60 USD |
| PBN, 100 slots | 100 USD | 9 | 900 USD |
| Twelve-month total | — | 12 | 6,357 USD |
Running the subtotals in order: 447 plus 4,500 is 4,947; plus 450 is 5,397; plus 60 is 5,457; plus 900 is 6,357. The two routes agree, which is the only thing this exercise is meant to establish.
Per click the maths fails, per contract it does not
Now attach outcomes to the year, still inside the same constructed example. Suppose the English site records 489 organic clicks across the twelve months, that 39 of those visitors send an enquiry — just under eight per cent, high because the traffic is specific — and that two of the 39 become orders.
| Denominator | Count in the year | 6,357 USD divided by it | Reads as |
|---|---|---|---|
| Organic clicks | 489 | 13.00 USD per click | Indefensible next to any consumer benchmark |
| Enquiries received | 39 | 163.00 USD per enquiry | Arguable, still the wrong unit |
| Orders won | 2 | 3,178.50 USD per order | The figure to take to the board |
Put the third row against the value of an order. At 85,000 USD of order value and a 35 per cent contribution margin, one contract contributes 29,750 USD. The full year of campaign cost, 6,357 USD, is 21.4 per cent of that single contract's contribution — so the campaign breaks even at roughly one-fifth of one order, and everything beyond that point is gain.
Cost per click
13 USD a click looks absurd against consumer benchmarks measured in cents, and comparing the two says nothing at all. The comparison is only meaningful between businesses with similar order values.
- Never present this figure alone
- It falls as traffic grows, which is not the point
Cost per contract won
3,178.50 USD to win an order contributing 29,750 USD is a ratio a finance director can evaluate in one pass, and it survives the traffic being small.
- Requires attribution discipline, not more traffic
- Ask what the same sum buys at a trade fair
Completing the year: two orders at 85,000 USD is 170,000 USD of order value and 59,500 USD of contribution. Subtract the 6,357 USD campaign cost and 53,143 USD remains. Expressed against order value, the campaign consumed 3.74 per cent of what it helped produce. Neither number is a prediction; both show which arithmetic answers the question a board actually asks.
Choosing between them without guessing
Five questions settle this more reliably than any feature comparison, because each maps to something the tiers genuinely differ on rather than to a preference.
- Is your vocabulary technical enough to be got wrong? Classification-society notations, tolerance classes and machine-specific terms are exactly where an automatic pool misfires, and that argues for manual selection.
- Can anyone internally review an on-site change? If nobody has the time or the standing to approve edits to a specification page, human-review mode is worth more than the price difference.
- Do you need writing, or only placement? The higher tier brings writers and developers; if you already have them, most of the gap is buying capacity you hold.
- How many domains are actually in scope? Pricing is per domain, so a portfolio of four country sites multiplies both tiers equally and changes the total rather than the choice.
- What does one order contribute? Where a single contract carries a five-figure contribution, the difference between the tiers is small enough that it should not be the deciding factor at all.
Three months on the automatic tier
Useful when nobody internally has run a campaign before and the first question is whether anything moves at all rather than which term moved.
- Log the start date and change nothing else
- Judge at ninety days, not at thirty
Straight to the reviewed tier
Sensible where the terminology is unforgiving, where pages carry regulatory content, or where there is simply no one to write the English pages.
- Set the DR threshold before placement begins
- Book a fortnightly slot for the approval queue
Questions that come up
Is the price per domain or per account?
Per domain, on both tiers. A company running a Finnish site and a separate English export domain pays twice, and a portfolio of country sites multiplies accordingly. The analytics and indexing surfaces are shared across everything connected to the account, but the campaign subscription attaches to a single domain.
Can we move between tiers, or add slots partway through?
The example above does exactly that, switching tier in month four and changing the PBN step at the same time. Budget it as two phases rather than one annual figure, and expect the month of the change to be the least readable month in the year, since two variables moved together.
Five hundred PBN slots cost 500 USD. Is that not obviously the best value?
No. The per-unit price is low because the asset is a different kind of thing from a manually placed link on a donor selected for its rating. Quantity at that step buys breadth of reference, not authority, and a link profile that does not match the size and specialism of the site is not an advantage. Judge the steps by what each type of placement is for.
Our search volume is nearly zero. Can a campaign be justified at all?
Only on contract value, never on traffic. If one order contributes tens of thousands and the year of campaign work costs a few thousand, the case rests on whether the work plausibly contributes to one additional order over a couple of years. That is a judgement about your sales cycle, and a volume-based business case will simply not close.
How soon should we expect to see anything?
Four to eight weeks is the usual window for first measurable movement, and on a low-volume export site the honest evaluation point is later still — ninety days at minimum, because a fortnight of data may contain three clicks. The campaign feed shows new placements with donor rating and traffic as they happen, which is activity rather than result.
Budget it as a plan, judge it as a contract
The two tiers differ in one axis: how much of the automation you take control of, and whether people come with it. 149 USD a month runs the campaign without asking you anything; 500 USD a month lets you choose the terms, aim the links, hold the edits, and puts specialists, developers and writers behind all three. Add-ons are per slot and sold in fixed steps, and the largest step is not automatically the best one.
The worked year came to 6,357 USD by two independent routes, which is the only claim this article makes about numbers. Everything attached to it — the clicks, the enquiries, the orders — was constructed to show which denominator survives contact with a business whose demand is measured in dozens and whose orders are measured in six figures.
Our other notes on measuring small markets sit with the rest of the English articles, and the way we set campaigns up alongside technical work is described on the service pages. Whichever tier you land on, log the start date and leave it alone for ninety days.
If you want the same arithmetic against your own domain rather than a constructed one, connect the property and let the panel gather a baseline before you commit to a tier: sign in and set up a campaign in the Semalt dashboard. On an export site the useful first question is never what the campaign costs a month. It is what one more order would have been worth.
