When should a single query get its own page?
When the answer this query wants is not the same answer as the overview page's topic, and you can write something other pages do not have. Split only when all three conditions hold:
1. The answer the query wants differs from the overview page
The overview page answers “which phrases are there”, while the query wants “which sticker to reply with when someone says thanks”; the shape of the answer differs.
2. The answer is buried in the overview page and hard to find
The overview page is long and this answer takes up only a small part of it, so both users and crawlers have trouble landing on it.
3. You have content unique to this query
Examples, steps, real data, usable assets; if you delete that query term, the page still has something useful left.
If any of the three conditions is missing, keep it in the overview page: add a subheading there, use the question as the heading, answer first and then expand; see How to write "X for Y" use-case pages that AI will cite for how to write it.
What does the real data of a single-query page look like?
MochiStamp’s data comes from Search Console records synced to the local dashboard (the operator’s own account, pulled 2026-11-12, window 2026-10-15 to 2026-11-11). mochistamp.com/fa/guide/reply-stickers is a Persian single-query page about “the sticker to send when replying thanks in a chat”; it is compared with the generic phrase pages in the same language, /fa/phrases/*, for example mochistamp.com/fa/phrases/teinei.
| Page | Data window | Impressions | Clicks | Click-through rate | Average position (impression-weighted) |
|---|---|---|---|---|---|
| /fa/guide/reply-stickers (1 page) | 2026-11-06 to 11-11, 6 days | 307 | 28 | 9.1% | 4.9 |
| /fa/phrases/* (7 pages) | Earliest 2026-10-30, to 11-11 | 603 (about 86 per page on average) | 28 | 4.6% | 6.4 |
In this one example the dedicated page does better, but this does not prove that a dedicated page is always better: the two sides went live at different times, the dedicated page has only 6 days of data, and all numbers are in the range of tens to hundreds of impressions. All we can say is that on this small site, a page written around a single query has higher impressions and click-through rate per page than the average generic page.
Which queries did this dedicated page capture?
The whole site has 8 queries of the “send a sticker in a reply” kind (Search Console syncs queries for the whole site, and our database has no mapping from a query to the page it landed on, so which ones landed on the reply-stickers page is inferred from the text and topic: unverified):
| Query | Impressions | Clicks | Average position |
|---|---|---|---|
| استیکر در جواب ممنون | 42 | 6 | 3.2 |
| استیکر در جواب مرسی | 40 | 7 | 3.6 |
| در جواب استیکر تشکر چه بگوییم | 32 | 2 | 6.6 |
| استیکر جواب تشکر | 19 | 0 | 3.7 |
| استیکر در جواب خواهش میکنم | 9 | 0 | 9.4 |
| استیکر ممنونم | 8 | 0 | 7.6 |
| در جواب استیکر چی بگیم | 3 | 0 | — |
| جواب مرسی به شوخی | 2 | 0 | — |
These 8 queries total 155 impressions, while the page's own impressions are 307; what queries make up the difference we do not know (Search Console does not show queries that appear very rarely): unverified. What these queries have in common is a single intent (which sticker to reply with after thanks), just worded differently, so one page is enough and there is no need for one page per wording. For how Persian and Arabic pages perform overall, see Arabic and Persian pages in search and AI answers: what one small site saw.
How do I keep a dedicated page and the overview page from competing for the same term?
1. One main query per page
The overview page covers the overview and the dedicated page covers that one query; the two pages do not use the same term in their title and H1.
2. The overview page points to the dedicated page with a descriptive link
Make the link text describe the dedicated page's topic, not “click here”; the overview page keeps only a sentence or two and this link, without repeating the dedicated page's content. For how to write internal links, see Do internal links help AI search?.
3. The dedicated page does not repeat the overview page's full list
Write only the answer this query wants and a few related examples; for everything else, the overview page links out.
4. After launch, check URLs by query
In Search Console, filter by query and check whether two URLs appear for the same query and which ranks first; our database has no such mapping, so we cannot currently see whether they compete: unverified.
5. Do not use canonical to merge two pages with different content
canonical means “these two URLs are the same content”; two pages with different content must be separated by division of labor, not by canonical; see Canonical tags and duplicate content: how to set them up on a multilingual site so pages are not treated as duplicates.
What is the difference between a single-query page and a page generated by swapping one word?
Google's spam policy (Google Search Central, original text read on 2026-11-12, the page shows no update date) says: “Scaled content abuse is when many pages are generated for the primary purpose of manipulating search rankings and not helping users.” It also says: “This abusive practice is typically focused on creating large amounts of unoriginal content that provides little to no value to users, no matter how it’s created.”
The policy's examples include: using generative AI or similar tools to produce many pages that add no value for users; scraping feeds or search results to generate many pages; stitching together content from different pages without adding value; and making many pages that mean nothing to readers but carry search keywords. The page gives no numeric threshold for how many pages or how similar counts as a violation, so the comparison below is our own practice, not Google's criteria:
| What it watches | Single-query dedicated page | Bulk page that only swaps one word |
|---|---|---|
| After you delete the query term | Useful examples, steps and data remain | Swap the word and the content stays the same; only the template remains |
| Where the content comes from | Written around the intent of this one query, with something unique to this page | The same template filled in with different words |
| Quantity | Based on whether there is real query demand, a few pages at a time | Many pages launched at once to cover a keyword list |
| Relationship to the overview page | Clear division of labor, linking to each other | Large blocks repeated between pages |
| After launch | Decide whether to keep or merge by impressions and clicks | Do not look at the data |
For whether thin-content pages should be noindex, see Should thin pages be noindexed? What it means for Google and AI search; for how to write “X used for Y” scenario pages, see How to write "X for Y" use-case pages that AI will cite.
How do we review the 4 new single-query pages later?
On the morning of 2026-11-12, MochiStamp launched 4 more single-query pages on the same approach: /fa/guide/khaste-nabashid, /fa/guide/congrats-stickers, /ar/guide/upset-stickers, /ar/guide/reply-stickers (we fetched these four URLs on 2026-11-12 and all returned 200). They have no data yet, so we will review them after two weeks:
1. Impressions and clicks
Impressions, clicks and click-through rate of each page, compared with the average of the generic pages in the same language.
2. Position
Average position, and whether the main query's position has stabilized.
3. Whether the overview page lost queries
After the dedicated page goes live, whether the overview page's impressions drop noticeably; if they do, the two pages are competing for the same term.
4. Decide to keep, rewrite or merge
For a page with no impressions, first check whether it has been indexed, then decide whether to rewrite it or merge it back into the overview page.
We draw no conclusion until the results of this section are in.
The queries, impressions, clicks and positions of single-query pages in Persian and Arabic are in Do Arabic and Persian long-tail pages work? Data from MochiStamp's reply-sticker pages.
How a query's average position can be a blend of a move partway through the period is in What does average position in Search Console mean? Position 8 but no clicks: how to read it.
FAQ
Is a query worth its own page?
Look at three things: the answer this query wants differs from the overview page's topic; the overview page is already long and this answer is buried in it; and you can write something other pages do not have. One Persian dedicated page on MochiStamp got 307 impressions and 28 clicks in the 6 days for which data is available; the 7 generic pages in the same language got 603 impressions and 28 clicks in total over 28 days. The numbers are small and this is only an example.
Is the dedicated page's data better than the generic pages'?
In this one example, impressions per page are higher (307 versus about 86 on average for the generic pages), click-through rate is higher (9.1% versus 4.6%), and average position is better (4.9 versus 6.4). But the two sides went live at different times and the numbers are small, so this does not prove a dedicated page is always better.
How do I keep a dedicated page and the overview page from competing for the same term?
One main query per page: the overview page covers the overview and the dedicated page covers that one query; the overview page links to the dedicated page with descriptive link text, and the dedicated page does not repeat the overview page's full list; after launch, filter by query in Search Console to see whether two URLs appear for the same query.
Can canonical solve two pages competing for the same term?
No: you cannot merge two pages with different content with canonical, because canonical tells search engines “these two URLs are the same content”. Two pages with different content must be separated by division of labor; see the canonical guide for how to write it.
How does a dedicated page differ from a page generated by swapping one word?
Look at what is left of the page after you delete that word. A dedicated page's content (examples, steps, data) is written around that query and does not hold up if you swap the word; a bulk page keeps the same content when the word changes. Google's spam policy calls many pages generated to manipulate rankings that have no value scaled content abuse, and gives no numeric threshold for page count or similarity.
What is Google's official position on bulk pages?
The policy's original text: Scaled content abuse is when many pages are generated for the primary purpose of manipulating search rankings and not helping users. Original text read on 2026-11-12; the page shows no update date.
How are the 4 new single-query pages doing?
No data yet. /fa/guide/khaste-nabashid, /fa/guide/congrats-stickers, /ar/guide/upset-stickers and /ar/guide/reply-stickers went live on 2026-11-12; we will review impressions, clicks and position after two weeks, and check whether the overview page lost queries.
Which queries did the reply-stickers page capture?
Our database has no mapping from a query to the page it landed on, so this can only be inferred from the query text: the whole site has 8 queries of the “send a sticker in a reply” kind, totaling 155 impressions, the largest being استیکر در جواب ممنون (42 impressions). Which of them landed on this page is unverified.
Sources
- Google Search Central · Spam policies for Google web search (scaled content abuse) (Original text read on 2026-11-12; the page shows no update date)
- MochiStamp · mochistamp.com/fa/guide/reply-stickers (Search Console figures from the operator’s own analytics) (The operator's own site; data window 2026-10-15 to 2026-11-11, pulled on 2026-11-12)
- MochiStamp · mochistamp.com/fa/phrases/* (Search Console figures from the operator’s own analytics) (The operator's own site; data window 2026-10-15 to 2026-11-11, pulled on 2026-11-12)
- kagentx evaluations: the data on the case study pages; each example gives its date.