Topic research based on your own site
No need to invent what to write about. The service scans your site, collects real queries from Wordstat, filters out what you have already covered and returns a list of topics — each can be taken into work with one click.
How the service picks topics
The order of steps is deliberate. The language model only does what it is good at: proposing what to ask statistics about and phrasing a headline. How many people search for a topic is answered by Yandex Wordstat, and the sums are calculated in code. Numbers never pass through the model, so the output contains no plausible but invented figures.
We look at your site
We read what is already published: via the WordPress API or the sitemap. The same data fills the internal linking index.
We identify the niche
We build the topic core from your headlines. This guards against homonymy: the same word can mean entirely different niches.
We ask Wordstat
For every seed we get hundreds of real queries with their volume — what people actually type into search.
We remove the noise
We cut out foreign niches, junk and your stop topics. Separately we drop what your site already covers — and show you exactly what.
We hand over the topics
Every topic comes with volume, intent, page type and a rationale. Accept it and it becomes a keyword ready for generation.
How this differs from just asking a neural network
Who needs topic research
Those just starting a blog
An empty keyword list is the first obstacle. Topic research gives a list you can start from right away, without studying Wordstat by hand.
Sites with a large blog
With a hundred articles it is hard to remember what is covered. The service checks topics against every page and shows only what is missing.
Agencies and contractors
A content plan for a new client is built in minutes instead of hours in Wordstat, and every number can be shown to the client with its source.
Those who do not do SEO
No need to understand search volume or clustering: the service does that work and explains why each topic was suggested.
Topic research in the service vs manually
| Parameter | Manually | SEO Writer |
|---|---|---|
| Collect queries for the niche | 1–2 часа в Вордстате | |
| Check what is already written | просмотр сайта вручную | |
| Filter out foreign niches | глазами по списку | |
| Group into topics | таблица и сортировка | |
| Source of search volume | Вордстат | Вордстат |
| Move into work | копирование по одному | одной кнопкой |
What users say
Real results from real people
“I used to pay 3–4 copywriters $15 per article and spend 2 days on approvals. Now I upload 20 keywords on Sunday — by Monday all articles are in WordPress. Organic traffic grew 40% in three months.”
“We manage 12 client sites. Content used to be the bottleneck — not anymore. SEO Writer generates texts with proper H1/H2/H3 structure, fills Yoast and internal links. Clients are happy, we can take on new ones.”
“We needed review articles for each product category. With SEO Writer we covered 60 queries in one month — with images, links to product pages and meta descriptions. That volume with copywriters would have cost $900.”
Frequently Asked Questions
Where does the search volume come from?
From Yandex Wordstat via the Yandex Cloud Search API. Every number in the interface is labelled with the region and measurement date, so a month later it is clear where it came from. The language model never sees volumes and cannot invent them.
Why is this better than asking a neural network for topics?
A language model is not connected to search statistics: it will propose plausible topics but cannot tell whether anyone searches for them, and it does not know what you have already written. Here the model only handles phrasing, while demand is confirmed by Wordstat and a comparison with your own site.
How does the service know what I have already written?
It scans the site: for WordPress it reads the post list via the built-in API, for others it takes URLs from the sitemap and fetches page titles. It also cross-checks the keywords you have already added in the service.
What if topics from another niche get suggested?
The niche core is built from your site headlines, and phrases that do not match it are dropped. Russian homonymy makes this mandatory: the same word can belong to entirely different niches. There is also a stop-topics field in the settings, so topics with those words will not appear.
Does topic research work for English-language sites?
Topics will be collected, but without volume: Wordstat only measures Russian-language demand. In that case we honestly mark the volume as unverified instead of substituting numbers from elsewhere.
How many runs are available?
The free plan includes one run per month with a shortened output so you can see how it works. Starter gets four, Pro ten, Agency thirty. The counter resets on the first day of the month.
Pricing
Start free, scale as you grow
Starter
1 290 ₽ · billed in RUB
- 15 articles/month
- 2 sites
- System API keys
- Topic research: 4 runs/month
- Custom API keys
- Scheduler
- Article rewrite for external platforms
Pro
2 990 ₽ · billed in RUB
- 50 articles/month
- 5 sites
- System API keys
- Topic research: 10 runs/month
- Custom API keys
- Scheduler
- Article rewrite for external platforms
- Cover generation
Agency
9 990 ₽ · billed in RUB
- 150 articles/month
- 20 sites
- System API keys
- Topic research: 30 runs/month
- Custom API keys
- Scheduler
- Article rewrite for external platforms
- Cover generation
- API access (coming soon)
- Priority support
Secure payment via YooKassa · No auto-renewal · 30 days access
Find out what your site should write about
First run free · Real Wordstat numbers · No credit card
Create free account →