# Publishing order: which pages of a map go live first

A map tells you which pages to build. The order tells you which to build first, and it is where most home-built plans fall apart. This lesson turns the rules people give into one order you can follow, with the evidence behind each step and a real branch from a client map.

By Kevin Jeppesen, AI growth engineer. Founder, FrameworkSEO & Ascendily. Updated 2026-11-13. Source: https://frameworkseo.com/learn/topical-maps/publishing-order

**Publishing order:** A publishing order is the sequence in which a map's pages go live, set so each page goes live after the pages it depends on and before the season it serves.

## Key takeaways

- Order matters because a new page starts with no record of how searchers react to it, and Google's court record shows ranking leans on that record: Navboost memorizes clicks over 13 months.
- Sort the rules into gates and priorities: a page goes live only when the page it feeds is live and its season leaves time, and only then do money, weak competition and search volume rank it.
- Never link to a page that is not live; write the words in plain text, put the date in your sheet, and add the link in both directions the day the target goes live.
- A 12-month average hides seasons: one winter question in a real map averages 1,300 searches a month but runs at 9,900 in January and 70 in summer, so it has to be live well before December.
- There is no controlled evidence for any pages-per-week number; the pace is yours to set, and every page still has to be done carefully, because a thin page does more harm than no page.

## Why the order you publish in changes what ranks

Order matters because a new page starts with no record of how searchers react to it, and Google leans on that record. Pages that go live early, and pages linked from pages people already visit, start building it sooner. The best evidence for this is Google's own words in court, not an SEO opinion.

In a December 2016 internal talk, filed as trial exhibit UPX0203 in the US antitrust case, Google engineer Eric Lehman put it bluntly: ["We do not understand documents. We fake it."](https://www.justice.gov/d9/2023-11/417516.pdf) The next slide says Google watches how people react and memorizes their responses. Another says the fakery fails "maybe because a document is new, recently-changed, or rarely-shown". A page you published this morning is exactly that document.

The court's 2024 ruling describes the system that does the memorizing. Judge Amit Mehta found that Navboost ["pairs queries and documents through memorizing user click data"](https://storage.courtlistener.com/recap/gov.uscourts.dcd.223205/gov.uscourts.dcd.223205.1033.0.pdf), trained on 13 months of data since 2017. Newer systems such as RankBrain and RankEmbed are, in the same ruling, "designed to fill holes in [click] data". They guess for pages with no history. They do not replace it.

The leaked 2024 API documentation adds one more piece. A field called [onsiteProminence](https://hexdocs.pm/google_api_content_warehouse/0.4.0/GoogleApi.ContentWarehouse.V1.Model.PerDocData.html) "measures the importance of the document within its site", computed "by propagating simulated traffic from the homepage and high craps click pages". Craps is the name the leaked documents use for click data. Its weight is unknown, but the idea is clear: a new page linked from pages people already reach looks more important than one linked from nowhere.

Koray Tuğberk Gübür, who named the topical map in his February 2021 case study, goes further. He argues that a page's first ranking reflects how much the engine already trusts the site on that subject. In his words, ["your first rankings affect your future rankings"](https://medium.com/@ktgubur/achieving-positive-ranking-state-120-growth-in-4-months-bcf51b590979) (March 2024). That is his reading of Google patents, not something Google has confirmed. The court record supports a weaker version, and it is enough to act on. Early pages start their record early. Pages with links from visited pages start it faster.

So the order is a set of bets on which pages should start earning first. How to read whether those bets are paying off is in [How to tell if a topical map is working](https://frameworkseo.com/learn/topical-maps/measure-a-topical-map).

## The six rules you will hear, and what each one protects

Six rules come up whenever people talk about order: money pages first, weakest competition first, hubs before spokes, connected batches, seasonal pages before their season, and fixes before new pages. Each one protects something real. Each one misleads when it is used alone.

| Rule | What it protects | Where it misleads, used alone |
| --- | --- | --- |
| Money pages first | Revenue: the pages that sell get their record started soonest | On a new site, the hardest searches go first and sit unseen for months |
| Weakest competition first | Early wins: pages that can reach page one quickly | Easy guides go live before the page they should send buyers to |
| Hubs before spokes | Structure: every guide has a live page to link up to | A hub can wait months for its own spokes if nothing else moves |
| Connected batches | Links: a hub and its pages go out together and link both ways | A batch that is never finished blocks everything behind it |
| Seasonal before the season | Timing: the page is found and tested before demand arrives | A seasonal guide jumps ahead of the service page it feeds |
| Fixes before new pages | Signals: an old page and a new one stop competing for one search | Fixing forever becomes a reason never to publish |

The guides that rank for this question each pick one rule and argue it. An agency guide published in 2026 says publish the pillar first, or in the same week as its first cluster page. A software vendor's guide says the opposite: publish three to five cluster pages first and the pillar after. A [2021 poll on Search Engine Roundtable](https://www.seroundtable.com/seo-content-priorities-30740.html) asked what a new site should build first: 41.7% of 672 votes said category pages, 26.3% said blog posts. Barry Schwartz's verdict was "it depends on the site".

It does depend, but not on taste. The contradictions go away once you see that the six rules do two different jobs.

![Six publishing rules in two groups: four gates that must be true before a page goes live, three priorities that rank it.](https://frameworkseo.com/learn/publishing-order/six-rules-two-kinds.webp)

*Gates decide whether a page may go live yet. Priorities decide which ready page goes next, and never open a gate.*

The money pages and the guides belong to different halves of the map. If the split between the pages that sell and the pages that explain is new to you, read [Core section and outer section](https://frameworkseo.com/learn/semantic-seo/core-and-outer-sections) first.

## How to turn the six rules into one order

Split the rules into gates and priorities. A gate must be true before a page may go live; a priority only ranks the pages whose gates are open. Then publish the highest-ranked ready page, and check again every time a page goes live.

This is not an SEO invention. It is how project schedules have been built for decades. List what each task depends on, and never start a task before its dependencies are done. Then choose among the tasks that are free to start by priority. Computer scientists call the result a topological order. A deadline is handled the same way, by working backward from the date. A map is a set of pages with dependencies, so it takes the same method.

### The gates

1. **Fixes before new pages.** If an existing page already chases a search the map gives to a new page, fix, merge or rebuild the old one first. Two pages for one search split what Google learns about both.
2. **The page it feeds is live.** A guide goes live only once the page it sends readers to exists. On an established site, the old version of that page counts.
3. **The season leaves time.** A seasonal page has a date it must beat, worked backward from the month its searches start to climb.
4. **The batch is complete.** A hub goes out with the pages it links to, so links can go in both directions on the same day.

### The priorities

1. **In the core, what the business most wants to sell.** The main service or collection first, then the rest in order of how much they matter to the business.
2. **Among guides, the weakest results first.** Count the forum threads, social posts and Q&A pages on page one for the guide's search. The more there are, the fewer good answers exist, and the sooner a clear guide can win. The how-to for this count is Step 8 of [How to build a topical map](https://frameworkseo.com/learn/topical-maps/how-to-build-a-topical-map).
3. **Searches a month, only to break a tie.** Volume never jumps a page past a gate.

![A three-step loop: clear existing pages, find pages whose gates are open, publish the best ready page, then repeat.](https://frameworkseo.com/learn/publishing-order/set-the-order.webp)

*The order is not set once. Each page that goes live opens the gates of the pages that feed it.*

### When two rules still disagree

Most fights between rules are a gate against a priority, and the gate wins. Weakest competition says publish the easy guide now; hubs before spokes says wait for its page. The guide waits. Seasonal timing says publish the winter guide in October; the gate says its service page is not live. Then the service page moves forward, so the guide can follow it in time.

The pillar-first and clusters-first camps are both partly right under this method. The clusters-first camp is right that small, winnable pages earn a record sooner. The pillar-first camp is right that those pages need somewhere to send readers. Ship the pillar with its first few cluster pages as one batch, and both get what they want. How a hub, its pages and their links go out together is in [Semantic content networks](https://frameworkseo.com/learn/semantic-seo/semantic-content-network).

## What to do with links to pages that are not live yet

Do not link to a page that is not live. Write the words in plain text, note in your sheet the date the target goes live, and add the link in both directions that day. A map is full of links between pages that go live months apart, so this comes up on almost every page.

There are two wrong ways to handle it, and one of them looks helpful.

- **A link to a page that does not exist yet.** Readers hit a dead end. Google's own advice is that a missing page does not hurt your other pages: in its 2011 post [Do 404s hurt my site?](https://developers.google.com/search/blog/2011/05/do-404s-hurt-my-site) it says the fact that some URLs return 404 "does not affect how your site's other URLs ... perform". It is still a broken promise to the reader. Koray lists checking that your pages do not link to non-existing pages as part of his clean-up before a launch (March 2024).
- **A "coming soon" page that stands in for it.** This is the worse one. A thin page that answers like a real page is what Google calls a soft 404. The same post says soft 404s "can have negative effects on our understanding and indexing of your site". You have published a page with no purpose, which is the opposite of what a map is for.

![A guide whose target is not live: no dead or coming soon link, plain text until launch, then links both ways on launch day.](https://frameworkseo.com/learn/publishing-order/links-before-live.webp)

*The link goes in on the day its target goes live, in both directions, never before.*

Koray worked this way in the case study that started the topical map. In his February 2021 write-up he says that on his test site ["we have put all the necessary internal links and anchor texts after the content is published"](https://www.holisticseo.digital/theoretical-seo/topical-authority). The links followed the pages, not the plan.

How often plans get this wrong has been measured once that I can find. A content agency audited its own calendar of 189 articles in 2026, before publishing any of them. 42 of the 173 cluster articles, 24%, were scheduled to go live before the pillar they supported, and the largest gap was nine months. The cause was sorting the whole calendar by keyword opportunity instead of by structure. That is a priority overruling a gate.

Which words to put on each link, and how many links a page should carry, are in [Internal links inside a topical map](https://frameworkseo.com/learn/topical-maps/internal-links-in-a-topical-map).

## How far ahead a seasonal page has to go live

A seasonal page has to be live, found and tested before its searches rise, not on the day they peak. Work back from the month the searches start to climb, and leave weeks for Google to crawl the page and settle its first rankings. Then check what that date means for the page it feeds.

The trap is the number in your map. Search volume is usually shown as a 12-month average, and an average hides a season. One row in the real map later in this lesson is the guide "How to Tell if a Pipe Burst in Your House". The map records 1,300 searches a month. The months tell a different story.

![US monthly searches for how to tell if a pipe burst: 9,900 in January 2026, 70 in summer, against a 1,300 average.](https://frameworkseo.com/learn/publishing-order/seasonal-lead.webp)

*One search, two seasons: 9,900 in January 2026, 70 in the summer, and a map average of 1,300.*

January 2026 had 9,900 searches, about 140 times the 70 of June, August and September. It is not one cold month either. In Google Trends over five years, the search peaked in December 2022, then in January 2024, January 2025 and January 2026, and fell to zero each July from 2022 to 2026. Pipes freeze in winter, and people search when the water appears.

How far ahead, then? Three dated facts set the floor, and none of them is a law.

- Google says crawling a new page ["can take anywhere from a few days to a few weeks"](https://developers.google.com/search/docs/crawling-indexing/ask-google-to-recrawl), and asking for it does not guarantee it (Search Central, updated December 2025).
- Koray launched his first content network on one site ["before the 45 days from the seasonal event"](https://www.oncrawl.com/technical-seo/creating-semantic-content-networks-with-query-document-templates-case-study/) it served, so the site could be tested against older competitors before demand arrived (Oncrawl, January 2022). That is his practice on one site, not a rule he proved.
- Navboost remembers 13 months of clicks, per the 2024 ruling. My reading, not Google's statement: a page that is live through one winter carries that winter's clicks into the next one. A page that misses a season does not just lose that winter. It starts its record a year later.

For the pipe guide, demand starts climbing in December. Koray's 45 days puts it live by mid-October, and the service page it feeds has to be live before that. If you only have the map's average, you would never move it.

> **Read the months before you trust the average:** Most keyword tools show the last 12 months behind the average. Look at them for every guide in your map. Any search whose busiest month is several times its quietest month gets a due date in your sheet. Check each row on its own: in the same branch, "burst pipe cleanup" peaked in July and August 2026, not in winter.

## What Koray Tuğberk Gübür says about momentum and pace, and what the evidence shows

Koray argues for a rising pace that never stops, and for holding back the most-searched topics until the site covers the ground around them. Google says publishing frequency itself is not a signal, and the only studies of pace are surveys and correlations. The order has reasons behind it. The pace mostly does not.

### What Koray did and says, with dates

- **A rising ramp (2021).** On the site in his first topical authority case study he published one article every three days, then one every two days after 20 days, then one a day after 10 more, then three a day after 20 more, counting language versions. He reports that Google gave the site more crawling after 26 August 2020 and reads that as Google noticing his consistency.
- **The biggest topics last (2021).** In the same study he writes: ["I haven't yet covered the most searched topics and most important topics in any way"](https://www.holisticseo.digital/theoretical-seo/topical-authority). His reasoning was that they would have had no visibility before the site had coverage around them.
- **Two speeds (2022).** On one project he published nearly 400 pieces of content in one day. On another he started with 10 to 15 and raised the speed steadily. Both are from his January 2022 Oncrawl case study.
- **Momentum (2023).** Of a site that stopped publishing, he wrote that ["momentum is lost"](https://www.holisticseo.digital/seo-research-study/semantic-search) and predicted it would lose traffic.
- **Clean up, then launch (2024).** His March 2024 case puts a technical sprint first, including cutting the total number of pages, and the new content network second.

His results are his reported numbers from his own screenshots, with no control sites. Treat them as strong anecdotes. More of his framework, in plain words, is in [Koray Tuğberk Gübür's topical map framework](https://frameworkseo.com/learn/semantic-seo/koray-topical-map-framework).

### What everyone else has

| Claim | Who | What it rests on |
| --- | --- | --- |
| Publishing more often is not itself a ranking signal | John Mueller, Google: ["A site isn't a machine that pumps out content at a fixed rate."](https://www.seroundtable.com/google-content-frequency-25367.html) (March 2018) | A Google statement, no data shown |
| No reason to hold back good pages | Mueller, on releasing 8,000 pages at once: ["why would you artificially delay it?"](https://www.seroundtable.com/google-delay-release-of-thousands-of-pages-35200.html) (April 2023) | A Google statement, no data shown |
| Many pages made for rankings, not readers, are spam | Google's [scaled content abuse policy](https://developers.google.com/search/docs/essentials/spam-policies) (updated August 2026) | A written policy |
| Effort per post goes with better results | Orbit Media's 2026 survey of 1,042 content marketers: spending 6 or more hours on a 2,000-word post roughly doubled the odds of reporting strong results | Self-reported results; "strong results" not defined; a correlation |
| Sites posting 9 or more articles a month grew organic traffic 20.1% | Stratabeat's 2025 comparison of 300 B2B software sites | A correlation; the method sits behind a form |

That is the whole evidence base. No controlled study compares one publishing speed against another on matched sites. Every "publish X a week" rule on the first page of Google rests on assertion or on one of the correlations above.

So here is how I put it to buyers who ask how fast to go. The pace is yours, and every page still has to be done carefully, because a thin page does more harm than no page. A gap in the schedule costs you a few weeks. A weak page costs you a page that ranks badly, drags on the pages around it, and has to be fixed or merged later. Koray's momentum point is worth keeping in its honest form. Once you start, do not stop for months. A site that goes quiet gives Google nothing new to learn from. That is a case for steady work, not for speed.

Why coverage over time builds trust on a subject at all is the argument of [Topical authority](https://frameworkseo.com/learn/topical-maps/topical-authority).

## How the order changes on a new site and on an established site

A new site starts with the pages that sell and has nothing to fix. An established site starts with what it already has: pages close to page one, pages that compete for one search, and old pages to rebuild before new ones join them.

### A new site

Put the core pages up before or at launch: the main service or collection, the other pages that sell, and the about and contact pages that prove a real business stands behind them. Then publish guides in batches behind the pages they feed. Hold the hardest head searches back, the way Koray held his most-searched topics back in 2021. A new site has no record to borrow, so its first guides should be ones with weak results, where a new page can actually be seen.

### An established site

Start in Search Console, not in the map. Three kinds of existing page go before any new one.

1. **Pages already near page one.** A page at position 11 to 20 for its search has the shortest way to go. Rebuilding it to the map is often the quickest win on the whole plan.
2. **Pages competing for one search.** Search engines limit how many results one site gets for a query. A 2011 Google patent application on this, filed as [US20110016108A1](https://patents.google.com/patent/US20110016108A1/en), describes engines that "limit the display of the results from a particular host". Bing's [Webmaster Guidelines](https://www.bing.com/webmasters/help/webmaster-guidelines-30fba23a) say duplicate URLs "dilute signals". Merge them before you add a third. See [Keyword cannibalization](https://frameworkseo.com/learn/topical-maps/keyword-cannibalization).
3. **Pages with no purpose.** Every page needs a reason to be on the site. A page without one is first merged into a more relevant page, and only removed if nothing fits. How to decide is in [Pruning and merging](https://frameworkseo.com/learn/topical-maps/pruning-and-merging).

One detail changes the gates on an old site: live is not the same as rebuilt. If the page a guide feeds already exists in its old form, the guide's gate is open, even if the rebuild of that page comes later in the order. You will see it in the real branch below.

> **Write down a before number:** In 2023 CNET deleted thousands of old articles to look "fresh" to Google, according to [Gizmodo](https://gizmodo.com/cnet-deletes-thousands-old-articles-google-search-seo-1850721475). Google's Danny Sullivan said the idea that Google dislikes old content is "not a thing", and no results were ever published. Before you fix, merge or rebuild a page, note its clicks and impressions for the last four weeks in your sheet.

## How the order changes for a store and for a service business

In a store the collections are the pages that sell, so they go first, and each guide follows the collection it sends readers to. In a service business the service pages take the calls, and the questions people ask right before they call come next.

| Decision | Online store | Service business |
| --- | --- | --- |
| Pages that go first | The main collection, then the collections that sell most | The main service, then the services that bring the jobs the business wants |
| Often already live | Collections and product pages: rebuild before adding | The service pages and a contact page: rebuild before adding |
| Guides that go early | Guides that end on one collection: sizing, fit, care, comparisons | The questions asked just before a call: cost, insurance, what happens next |
| What sets the seasons | Gifting dates, weather, the store's own sales calendar | Weather and emergencies, plus the local calendar (tax dates, school terms) |
| The usual mistake | A big blog guide goes live before the collection it should sell | Location pages swapped city by city, with nothing true only of that town |

Two real maps show the pattern. In a 391-page store map opened up in [Topical map examples](https://frameworkseo.com/learn/topical-maps/topical-map-examples), the main collection is the first page in the order, and the biggest guide in the branch shown there goes live 145th, after the collection it links to. In the service map below, the cost question and the insurance question go live at 10th and 8th, ahead of the repair service at 15th.

The store version of the full method is in [Topical maps for ecommerce stores](https://frameworkseo.com/learn/topical-maps/topical-map-for-ecommerce), and the service and local version in [Topical maps for local and service businesses](https://frameworkseo.com/learn/topical-maps/topical-map-for-local-and-service-businesses).

## A real branch in publishing order, read rule by rule

Ten rows from a real 173-page map for a US water and fire damage restoration company show the rules at work: rebuilt service pages first, every guide after the page it feeds, and one winter question that the calendar, not the map, has to move. The business name and its state are replaced with [brand] and [state]. No other word in the rows is changed.

P is the page's place in the order across the whole map, so the numbers jump. The last column is not from the map: it is the month each search peaked in the US between November 2025 and October 2026, pulled for this lesson.

| P | Page | Type | Feeds | Existing page | US a month | Peak month |
| --- | --- | --- | --- | --- | --- | --- |
| 3 | Water Damage Restoration in [state] \| [brand] | service (hub) | Water, Fire & Storm Damage Services \| [brand] | Yes | 90,500 | Dec 2025 |
| 4 | Water Damage Cleanup in [state] \| [brand] | service | P3 | Yes | 2,900 | Oct 2026 |
| 5 | Flooded Basement Cleanup in [state] \| [brand] | service | P3 | No | 1,600 | Jul and Sep 2026 |
| 6 | Burst Pipe Water Damage Cleanup in [state] \| [brand] | service | P3 | No | 1,000 | Jul and Aug 2026 |
| 8 | Does Homeowners Insurance Cover Water Damage? | blog | Water & Fire Damage Insurance Claim Help \| [brand] | No | 1,600 | Oct 2026 |
| 10 | Water Damage Restoration Cost: What Drives the Price | guide | P3 | No | 720 | Feb and Mar 2026 |
| 15 | Water Damage Repair in [state] \| [brand] | service | P3 | No | 9,900 | Oct 2026 |
| 19 | Water Mitigation Services in [state] \| [brand] | service | P3 | No | 5,400 | Oct 2026 |
| 32 | How to Tell if a Pipe Burst in Your House | blog | P6 | No | 1,300 | Jan 2026 |
| 127 | Dehumidifier for Water Damage: How Long to Run It | blog | P19 | No | 6,600 | Jan to Apr 2026 |

![A real water damage branch as a tree: service pages from P3 to P19, and each guide numbered after the page it feeds.](https://frameworkseo.com/learn/publishing-order/real-branch-order.webp)

*Every guide's number is higher than the number of the page it feeds. The gate holds across the branch.*

### Reading it with the rules

- **Fixes first.** P3 and P4 are pages the site already has. They are rebuilt before any new page in the branch goes live.
- **The gates hold.** P10 feeds P3, P32 feeds P6, P127 feeds P19. Each guide comes after its page.
- **Live is not rebuilt.** P8, the insurance question, feeds a claim help page the site already had. Its gate is open from day one, so it can go early.
- **Priorities inside the gates.** The cost guide at P10 goes before the repair service at P15, though repair has 9,900 searches a month against 720. My reading: its page is live, and it answers what people ask right before they call. Volume does not set the order.
- **The season is yours to apply.** P32 peaks in January. A map's order is the order for an ordinary month. Only you know the date you start, so you set the due date. If your start means P32 lands after December, move P6 and P32 forward as one pair, as far as you can do both properly.
- **Big does not mean early.** The dehumidifier guide has 6,600 searches a month and goes live 127th. Its page, water mitigation, comes at P19, and many other pages of the map come before it.

## The schedule sheet to keep, and what to do when a page stalls

Keep one sheet with a row per page: its gate, its due date, the planned and actual live dates, the links added, and what it showed after a few weeks. It holds the order and the record in one place, and it is the only way to know later what worked.

[Download the publishing order schedule (CSV)](https://frameworkseo.com/learn/publishing-order/publishing-order-schedule.csv) (17 columns with the ten real rows from this lesson filled in, their peak months, and empty columns for your dates, links and results. Opens in Google Sheets or Excel.)

The columns, and why each one is there:

- **Gate: must be live first.** The page this one feeds. If that page is not live, this row waits.
- **Peak month and season due date.** From the monthly numbers, not the average.
- **Planned and live dates.** The gap between them is your real pace. Plan the next month with that number, not with a wish.
- **Links added to it and from it.** The day a page goes live, open every live page that should link to it, add the link, and tick it here. Then add its own links to the pages that are already live.
- **Submitted.** Inspect each new URL in Search Console. For Bing, Yandex, Seznam, Naver and the other engines in [IndexNow](https://www.indexnow.org/documentation), one request can list up to 10,000 URLs, so a whole batch goes in one call. Google is not part of IndexNow.
- **Impressions after 4 and 12 weeks.** Impressions come before clicks. A page that is indexed and gaining impressions is working, even before it brings visits.

### When a page stalls

A page that shows almost nothing after it has been live for a few weeks gets three checks, in this order.

1. **Is it indexed?** URL Inspection in Search Console answers this. A page Google has not indexed cannot rank, and the cause is usually links or quality, not order.
2. **Do live pages link to it?** Check the links-added column. A page nobody links to has only its own record to work with.
3. **Does it answer its search better than page one does?** Read the results again. If they changed since you planned the page, the page may need a new angle, not more time.

How long to wait and what normal growth looks like is in [How to tell if a topical map is working](https://frameworkseo.com/learn/topical-maps/measure-a-topical-map). Once the first pages are live and you can see what Google rewards, the map itself can grow; that is [Expanding a topical map](https://frameworkseo.com/learn/topical-maps/expanding-a-topical-map).

Everything in this lesson works by hand with a spreadsheet and Search Console. If you would rather have the whole map built for your site, with every page already in its order, that is our [growth plan](https://frameworkseo.com/).

## Sources

- [Importance of Topical Authority: A Semantic SEO Case Study](https://www.holisticseo.digital/theoretical-seo/topical-authority), Koray Tuğberk Gübür, Holistic SEO, 2021
- [Creating semantic content networks with query and document templates: a case study](https://www.oncrawl.com/technical-seo/creating-semantic-content-networks-with-query-document-templates-case-study/), Koray Tuğberk Gübür, Oncrawl, 2022
- [Semantic search and frame semantics case study](https://www.holisticseo.digital/seo-research-study/semantic-search), Koray Tuğberk Gübür, Holistic SEO, 2023
- [Achieving Positive Ranking State: 120% Growth in 4 Months](https://medium.com/@ktgubur/achieving-positive-ranking-state-120-growth-in-4-months-bcf51b590979), Koray Tuğberk Gübür, Medium, 2024
- [UPX0203: Q4 Search All Hands (trial exhibit, US v. Google)](https://www.justice.gov/d9/2023-11/417516.pdf), US Department of Justice, 2016
- [Memorandum opinion, US v. Google LLC, No. 20-cv-3010 (D.D.C.)](https://storage.courtlistener.com/recap/gov.uscourts.dcd.223205/gov.uscourts.dcd.223205.1033.0.pdf), US District Court for the District of Columbia, Judge Amit P. Mehta, 2024
- [PerDocData (leaked Content Warehouse API documentation)](https://hexdocs.pm/google_api_content_warehouse/0.4.0/GoogleApi.ContentWarehouse.V1.Model.PerDocData.html), hexdocs.pm copy of the 2024 leak, 2024
- [Search result plusbox including restricted results (US20110016108A1)](https://patents.google.com/patent/US20110016108A1/en), Google LLC, patent application, 2011
- [Bing Webmaster Guidelines](https://www.bing.com/webmasters/help/webmaster-guidelines-30fba23a), Microsoft Bing, 2026
- [IndexNow documentation](https://www.indexnow.org/documentation), IndexNow.org, 2026
- [SEO poll: which content should SEOs prioritize](https://www.seroundtable.com/seo-content-priorities-30740.html), Search Engine Roundtable, 2021
- [CNET is deleting thousands of old articles to game Google Search](https://gizmodo.com/cnet-deletes-thousands-old-articles-google-search-seo-1850721475), Gizmodo, 2023
- [Google: content publishing frequency not a ranking signal](https://www.seroundtable.com/google-content-frequency-25367.html), Search Engine Roundtable, 2018
- [Google: no SEO reason to delay release of thousands of pages](https://www.seroundtable.com/google-delay-release-of-thousands-of-pages-35200.html), Search Engine Roundtable, 2023
- [Do 404s hurt my site?](https://developers.google.com/search/blog/2011/05/do-404s-hurt-my-site), Google Search Central, 2011
- [Ask Google to recrawl your URLs](https://developers.google.com/search/docs/crawling-indexing/ask-google-to-recrawl), Google Search Central, 2025
- [Spam policies for Google web search](https://developers.google.com/search/docs/essentials/spam-policies), Google Search Central, 2026