In 14 Romanian niches, measured with the same tools on the same day, the organic window is open: of 68 probed queries, 64 sit under difficulty 15 — and in 10 of the 14 niches, all of them do. The gap is not on the generic term, where retailers and aggregators sit. It is on local ("in my city"), on procedure ("how long, how much does it hurt, what do I need to do") and on price ("how much does it cost"). Below: the numbers per niche, the four patterns that come out of them and — just as important — what these numbers do NOT say.
How we measured (so you can repeat it)
This is not an opinion poll and not a purchased report. On 17 August 2026 we took 14 niches in which we have no clients, two search "seeds" per niche, and extracted with Ubersuggest (Romania database) the representative queries: estimated monthly volume and SEO difficulty (SD) — the tool's score for how hard it is to enter the top 10, from 0 to 100. We kept 2-7 queries per niche, 68 in total: local, procedural, price and the main generic term as a reference.
The threshold we use internally: SD under 15 = weak competition in the top 10, by the tool's metric. It is not a guarantee of a position — it is the sign that nobody serious has built there yet.
The numbers, niche by niche
| Niche | Queries probed | Under SD 15 | SD band | Cumulative volume (monthly, est.) | Where the gap is |
|---|---|---|---|---|---|
| Dentistry | 6 | 6 | 5–9 | ~1,000 | procedural + local + price |
| Accounting | 3 | 3 | 6–14 | ~360 | post-incorporation |
| Legal services | 2 | 2 | 7–8 | ~280 | secondary cities |
| Veterinary clinics | 6 | 6 | 6–10 | ~3,900 | local, city by city |
| Wedding venues | 4 | 3 | 7–18 | ~1,900 | large cities (the generic is above 15) |
| Solar panels and heat pumps | 6 | 6 | 8–12 | ~9,800 | kit + price + installation |
| Car service and parts | 4 | 4 | 7–13 | ~7,900 | 24/7 + local + labour rates |
| Car detailing | 6 | 6 | 4–13 | ~6,100 | local + price + at home |
| Laser hair removal | 5 | 5 | 6–14 | ~4,600 | local, concentrated in 6 cities |
| Perfumes online | 7 | 6 | 5–31 | ~38,400 | flanks: decants, minis, niche (the generic belongs to retailers) |
| Home & deco | 4 | 3 | 11–15 | ~1,450 | material + style, long tail |
| Kids and babies | 4 | 4 | 11–13 | ~780 | comparisons + safety |
| Pet shop online | 5 | 5 | 10–14 | ~2,250 | local + product guides |
| Supplements | 6 | 5 | 11–27 | ~36,700 | by symptom, not by brand |
Source: Ubersuggest, Romania database, 17 August 2026. Volumes are the tool's estimates, rounded; difficulty is its score, not Google's.
The four patterns that come out of the numbers
The gap is local, not generic
For vets, "veterinary clinic" has over 12,000 searches a month nationally — and a difficulty not worth the fight. But the city variants, hundreds of them, sit at SD 6-10 with 500-900 searches each. The same picture for car service, laser hair removal and detailing: whoever builds the honest page for their own city enters an almost non-existent competition.
The procedural question beats the brand
In dentistry, the queries that rank easily are not about clinics — they are "how long does the pain last after an implant", "how long does it take to place an implant", "which vitamins after an implant". In accounting: "obligations after incorporating a company". In supplements: not the brand (saturated, heavy ads), but the symptom. Whoever answers the question that comes before the purchase gets in front of the buyer before the seller does.
The big generics belong to the aggregators — and that is fine
"Men's perfumes" has 27,100 searches and SD 31; "Arabic perfumes", also 27,100 and SD 57. You do not go in head-on there. But "perfume decants" — 1,600 searches, SD 5 — is a flank the big retailers have not covered. The rule: you do not attack the mountain, you attack the pass.
High volume does not mean high difficulty
The prejudice: "if it has many searches, it must be hard". Solar panels contradict it: "5 kW solar kit price" has 2,900 searches a month at SD 12. Consolidated transactional demand, weak content competition — because the industry sells, but does not explain.
What these numbers do NOT say
We like publishing numbers, but only with their limits written next to them:
- SD is a tool's metric, not Google's. A low score says the pages in the top 10 are weak by the tool's criteria — not that Google will put you there.
- Volumes are estimates. Rounded, with their margin; the direction matters, not the third digit.
- An easy query is not a customer. Intent, the page that receives the visitor and trust make the difference between traffic and money. We did not measure conversion — we do not promise it.
- 14 niches are not the market. They are the ones we chose and could measure completely on the same day. Your niche may look different; the method is the same and takes a day.
- An open window closes. The numbers are from August 2026; the first agency that builds seriously in a niche changes its difficulty for everyone else.
What to do with this, if you run a business in one of these niches
You do not buy a keyword list — the list ages and teaches you nothing. You build, in this order: the page for your city, written for people (how local SEO is done without looking like a doorway), then the answers to the procedural and price questions, linked to each other and to the service page (the octopus model). And if you want to see the numbers on your own site, not on the niche in general, the engineering audit measures yours.
FAQ.PROTOCOL
Frequently Asked Questions
Let's build something remarkable.
30-min discovery call — no cost, no pitch. We audit your digital architecture and deliver a clear operational plan.
- 01Short message with your business context
- 02Reply within 24h with a discovery-call proposal
- 03Operational plan + scope recommendation