Changelog

The homepage now leads with the site's own name.

Over the last ninety days every single query that reached the homepage was the site's name or a misspelling of it — "seo studio tools", "seo studio", "seostudio tools" and so on. Not one generic search landed there. The page was still half-written for generic "YouTube SEO tools" language, which is a job the guide at /youtube-seo-tools took over. So the homepage has been re-pointed at what people actually type to find it.

• The headline now names the site: "SEO Studio Tools — free YouTube SEO tools backed by real trending data". Before it did not contain the name at all.
• The title tag and the search-result description both lead with the name and say plainly that the tools are free and need no sign-up.
• The first sentence on the page now states what SEO Studio Tools is and who publishes it, which is the sentence search engines read to work out what an unfamiliar name refers to.
• A short "What this site is" section has been added near the top: what the tools do, that every suggestion carries a measured count from the daily index, that nothing needs an account, and — just as important — what the site cannot do.
• The structured data now declares the site's alternative spellings and lists the main destinations, which is the material Google uses to build the extra links that appear under a brand result.

Nothing else moved. The advertising unit, the tools grid, the live index figures, the trending preview and every other page's titles and descriptions are untouched.

Pages changed: /, /changelog

The single advertising unit is switched on.

• One responsive banner now runs directly under the header, above the breadcrumb and the headline, on every public page and on the older tools. There is no second unit anywhere, and none at all on the sign-in and sign-up pages.
• It spans the content column, so its left and right edges line up with the text below it rather than running edge to edge across the screen.
• Its space is reserved before it loads — 101 pixels on a phone, 91 on a desktop — so the page does not move when the ad arrives, or when it does not. Measured across 72 combinations of width, theme and page: nothing shifts, and layout shift stays at 0.015 in Lighthouse, which is Google's own consent dialog rather than the page.
• The publisher and slot ids now live in one configuration file instead of being typed into two templates, so there is a single place to change them.
• The console is clean. Google's loader inserts a tiny invisible marker of its own, and the guard that strips unwanted ad containers was removing it, which made the loader log an error on every page. It is now left alone, and Lighthouse's best-practices score went from 96 to 100 as a result.

The unit may show blank for a day or two. A newly created ad slot takes a while before AdSense starts filling it, and the reserved space means nothing on the page moves when it does.

Pages changed: /, /trending-tags, /trending-tags/india, /youtube-tag-generator, /youtube-title-generator, /changelog, /methodology, /api

Four pages rewritten and one new guide, driven by what people are actually searching for.

Find Facebook ID - The page was a bare input box with a few lines of copy. It is now a full answer: what a Facebook ID is and how it differs from a username, every URL form the tool accepts and what each resolves to, and five manual methods with exact click paths for people who would rather not use a tool at all. - It says plainly that Facebook serves a login wall to servers, so most personal profiles cannot be resolved automatically, and gives the manual workaround instead of returning a number that might be wrong. - New sections on reading each failure message, and on what a Facebook ID is not useful for — it reveals nothing private, it cannot recover an account, and it cannot find someone by name. - Ten questions answered properly at the bottom.

A new guide: YouTube SEO tools - A new page at /youtube-seo-tools explaining what each category of YouTube SEO tool actually does, what it is worth, and where it stops measuring and starts guessing. It names the real tools — TubeBuddy, vidIQ, Keyword Tool, RapidTags, YouTube Studio, Google Trends — and says what each is genuinely good at. - It states the thing most pages in this category avoid: nobody can show you real YouTube search volume, because YouTube does not publish it to anyone. Every figure you see elsewhere is modelled. - It carries live numbers from the index, so it is the only guide of its kind that refreshes daily.

YouTube Description Extractor - Rewritten to cover what people call it — viewer, reader, grabber, checker — and to explain what the description field is actually worth next to the title and the thumbnail, why the first 157 characters carry the weight, and exactly which videos will not extract and why.

YouTube Channel Logo Downloader - Rewritten to cover logo, icon, avatar and profile picture as the same image, the sizes YouTube serves and which to pick for what, and every way you can identify a channel. - A prominent, honest section on copyright: someone else's channel logo is their work and usually their trademark, and downloading it is not a licence to use it.

Fixes along the way - Several tool pages showed YouTube furniture regardless of what the tool did. Find Facebook ID, which has nothing to do with YouTube, was claiming its evidence came from the YouTube Trending Tags Index and listing YouTube tools beside it. Every tool now declares where its data comes from, and the page furniture follows. - The language switch on the methodology, changelog, API and new guide pages pointed at Arabic versions that do not exist. Those four pages no longer offer it. - A long URL in a worked example pushed the channel logo page sideways on a phone. Long links in code now wrap.

Pages changed: /find-facebook-id, /youtube-seo-tools, /youtube-description-extractor, /youtube-channel-logo-downloader, /youtube-tag-generator, /youtube-title-generator, /methodology, /api, /changelog

The daily fetch had not run since 4 September. The index is current again, and the job that keeps it current is now actually installed.

The index - The trending index was two days stale. The cause was a missing cron entry: the scheduler that runs the daily fetch had never been installed for the site account, so the job registered inside the application was never triggered. It is installed now and runs every minute; the fetch itself runs at 03:00 UTC. - All eight countries were refetched. India, the United States, Pakistan, Indonesia, Bangladesh, Brazil, Egypt and Nigeria are all current, and the country cards were redrawn with the new numbers. - If a country ever falls more than 36 hours behind again, its page and the homepage say so in plain words — "Last fetch was 2 days ago, the daily job did not run" — and a warning goes to the log. The timestamp shown is always the real one; nothing rounds a stale fetch up to today. - There is a new command, youtube:index-health, that prints the last fetch, its age and the row counts for every country in one table.

What else changed - The blank advertising band is gone. With no ad slot configured the site was reserving a 100-pixel white strip under the header on every page, pushing everything down for nothing. It now renders nothing at all until an ad slot is set. - With that strip gone and the hero tightened, the headline on the homepage now starts 136 pixels down a phone screen instead of 313, so the opening lines and both buttons are visible without scrolling. - The cookie notice was rebuilt. It holds its own space open at the foot of the page so it can never cover a button, it closes with the Escape key, its buttons are full-size tap targets, and it reads correctly in dark mode and in Arabic. - The sign-in, sign-up and password pages asked search engines to index them. They are now marked noindex, which is what an account page should be. - "Distinct channels" on a country page and the number of trending videos are now visibly different figures. They always were counted separately, but with only one day of data in the window every video happened to come from a different channel, so the two numbers matched and looked like a bug.

Pages changed: /, /trending-tags, /trending-tags/india, /trending-tags/united-states, /trending-tags/pakistan, /trending-tags/indonesia, /trending-tags/bangladesh, /trending-tags/brazil, /trending-tags/egypt, /trending-tags/nigeria, /changelog

A design QA pass over the redesign: every page photographed at six widths in both themes, every capture measured, and the twenty-nine defects that turned up fixed. Advertising was cut back to a single unit.

Advertising - The site now runs exactly one ad: a responsive display unit directly under the header, on every page. The five in-content slots and three sidebar slots the original script shipped are gone. - The band that holds it reserves its height before the ad arrives — 101 pixels on a phone, 91 on a desktop — and the unit sits out of flow inside it, so an arriving or missing ad cannot move the page. Measured across 252 page states with the ad script running: nothing moves at all. Before this change a phone measured 0.715. - Anything the ad network tries to place on its own — an extra unit, an auto-placed slot in the header or the footer — is removed before it fills. A filled ad is never touched. - The header is capped at 72 pixels on a phone and 88 on a desktop so nothing injected can stretch it again.

What was wrong and is now fixed - Text and buttons jumped about a second into every page load. The web fonts were set to swap in when they arrived, and two of the four the first screen uses were not being fetched early, so the page was re-laid-out under the reader. The fonts now commit once and never swap, and all four are fetched up front. The worst page measured 0.106 of movement before and none after. - The mobile menu did not move focus into itself when it opened, let focus escape while it was open, and did not hand focus back on Escape. All three are fixed, and the hamburger no longer draws its open and close icons on top of each other. - The homepage claimed 17,031 distinct tags. The real number across all eight countries is 13,747 — the old figure added up the per-country counts, so a tag trending in three countries was counted three times. - The sign-in and sign-up pages were still wearing the old theme: a purple panel, no heading, an unlabelled email field, a second page landmark inside the first, and web fonts fetched from Google on every visit. They are now on the same design as the rest of the site. - An unfilled ad painted a white rectangle across dark pages. - Headings dropped their last word onto a line of its own on the privacy page and the not-found page. Headings now balance their lines. - A heading in the tool sidebar was rendering at 12 pixels; one paragraph of tool copy ran past the reading measure; the narrowest phone had to scroll a four-column table sideways; and the main text box on the older tools had no accessible name for a screen reader. - The admin "Edit page" button sat above the headline in the content. It is now a small pill in the corner, and it was never in the HTML sent to visitors. - "Channels appearing" on a country page showed the same number as "videos measured", which read like a bug. It was not — the chart rarely repeats a channel — so the figure is now labelled "Distinct channels" and says so in a line underneath. Egypt, where eight channels do repeat, shows 203 of 211.

Every title, description, robots rule, canonical and structured-data block is unchanged, and the sitemap still lists the same 29 pages.

Pages changed: /, /about-us, /contact, /methodology, /changelog, /api, /trending-tags, /youtube-trend, /youtube-title-generator, /youtube-tag-generator, /youtube-hashtag-generator, /youtube-tag-extractor, /youtube-tag-gap, /youtube-description-extractor, /youtube-thumbnail-downloader, /youtube-channel-id, /youtube-channel-logo-downloader, /youtube-money-calculator, /find-facebook-id, /trending-tags/india, /trending-tags/united-states, /trending-tags/pakistan, /trending-tags/indonesia, /trending-tags/bangladesh, /trending-tags/brazil, /trending-tags/egypt, /trending-tags/nigeria

The site had the same face as every other install of the script it is built on: a purple gradient hero, rainbow tool cards and stock icons. The product underneath is not the same thing any more, so the design is not either.

What the site looks like now - A new identity: the mark is a tag with three ascending bars, drawn once and used everywhere from the 16-pixel favicon to the social cards. Deep green means one thing on this site — measured evidence — and amber means one thing: untested. There is no purple and no gradient anywhere. - Inter for the interface and JetBrains Mono for every number, tag and table cell, both self-hosted rather than fetched from a third party on each visit. - Light and dark themes, switchable from the header and remembered only if you actually choose one. - Evidence now reads as one thing: IN THE INDEX, then the video count and the average views, in mono green, right under the tag it belongs to.

What was rebuilt - One hand-written design system replaced the 352 KB theme. Kept pages now load 10 KB of compressed CSS and 2 KB of JavaScript instead of Bootstrap, jQuery, Popper, Pace, two scrollbar libraries and an icon font. - The homepage leads with the index: a live strip showing videos measured, countries, distinct tags and the time of the last fetch, then the twelve tools, then the ten tags trending in India right now. - Tool pages are two columns on a desktop: the tool on the left, "how it is measured" and the related tools on the right. The old Popular Tools and Recent Posts widgets are gone. - Country pages read as data pages: stat tiles, a sticky action bar with the CSV and JSON buttons, tables with sticky headers and right-aligned mono numbers, hashtags as click-to-copy chips. - The API docs show a response taken from the live index at the moment the page is rendered, not a paste that will rot. - The ~160 general-purpose utilities inherit the new design automatically. Nothing about them was redesigned by hand.

What was fixed - Ads now sit in slots that reserve their height before the ad loads. Layout shift on a tool page fell from 0.47 to 0.015. - The theme used to load a 275 KB hero image on every page and Google Fonts at runtime. Both are gone. - Every colour pair in both themes was measured against WCAG AA; three tokens were darkened until they passed on every background they are used on. - Keyboard users get a skip link, a visible three-pixel focus ring on every control, and a full run through any tool without a mouse. - The Arabic pages mirror properly: the layout uses logical properties throughout, so nothing overflows and the logo does not flip. - Social cards are generated per page, and the eight country cards are rebuilt after every daily fetch so they always show that day's numbers.

Pages changed: /, /about-us, /contact, /methodology, /changelog, /api, /trending-tags, /youtube-trend, /youtube-title-generator, /youtube-tag-generator, /youtube-hashtag-generator, /youtube-tag-extractor, /youtube-tag-gap, /youtube-description-extractor, /youtube-thumbnail-downloader, /youtube-channel-id, /youtube-channel-logo-downloader, /youtube-money-calculator, /find-facebook-id, /trending-tags/india, /trending-tags/united-states, /trending-tags/pakistan, /trending-tags/indonesia, /trending-tags/bangladesh, /trending-tags/brazil, /trending-tags/egypt, /trending-tags/nigeria

The site is now built around one thing it owns: a daily index of the tags, hashtags and title patterns behind the videos trending on YouTube in eight countries. Everything else was either rebuilt on top of that index or taken out of the way.

What was added - The YouTube Trending Tags Index at /trending-tags, with a page per country for India, the United States, Pakistan, Indonesia, Bangladesh, Brazil, Egypt and Nigeria. Each one carries the top 100 tags, the top 50 hashtags, measured title patterns, the channels appearing most often, a CSV download and the exact timestamp of the fetch it is showing. - A daily job that pulls up to 200 trending videos per country from the YouTube Data API and rebuilds the index. It runs at 03:00 UTC and costs about 32 quota units a day. - A new tool at /youtube-tag-gap: paste your channel and a competitor's, and see every tag they use on their 50 most recent uploads that you do not, ranked by the views behind it. - A free JSON API at /api with five endpoints, no key and no sign-up. - A methodology page at /methodology that spells out where every number comes from and what these tools cannot do. - This changelog. Any "last changed" date shown anywhere on the site is now read from it.

What was rebuilt - All eleven kept YouTube tools now read live data through the YouTube Data API v3, accept every URL form including Shorts links and @handles, and cache each lookup for 24 hours. - The tag, hashtag and title generators now show evidence next to every suggestion: how many videos trending in your country this week use it, and what those videos average in views. Suggestions with no evidence are labelled untested rather than mixed in. - The title generator gained a Title Score that benchmarks a title against the titles actually trending in the country you pick, with the measured share behind every factor. - The extractors accept up to 20 URLs at a time with a CSV export, with no login. A per-IP hourly limit replaced the login wall. - Every kept tool page lost its generic 14-heading essay and gained specific copy: what the tool does, how to use it, what it cannot do, real questions, and a worked example rendered from the live index. - The Find Facebook ID tool now resolves pages and groups through Facebook's public page plugin, and says plainly when Facebook's login wall blocks a personal profile instead of returning a wrong number.

What was fixed - /about-us returned 404 while the footer linked to it. It now exists, names the publisher (Thinkly for Digital Business, Egypt), gives a real contact address and states an editorial policy. - The contact page shows the publisher and contact address above the form. - The homepage had the meta description "Seo Studio Tools". It now has a real one, a real title, and Organization structured data. - 202 template pages that were identical to every other install of this script were set to noindex and dropped from the sitemap. They all still work and are all still linked from the homepage. - The /ar mirror is noindex and out of the sitemap. hreflang tags were removed everywhere, because there was no complete alternate set to declare. - The sitemap now lists 29 URLs instead of every page in both languages, and every lastmod comes from a real change rather than the time of the request. - The blog link was removed from the navigation and footer; there is one English post and it is still reachable. - Kept pages defer their JavaScript and load web fonts without blocking the first paint.

Housekeeping - Contrast, tap-target and heading-order problems in the shared header, footer and result tables were fixed, so the rebuilt pages now score 100 on Lighthouse accessibility and best practices. - The hero image was 275 KB and loaded lazily even though it is the first thing you see. It is now served as WebP or JPEG at the size your screen needs, and loaded first. - Stylesheet and script URLs now carry a version stamp, so an edit is visible immediately instead of waiting for the CDN cache to expire. - The server was willing to serve dotfiles, .bak files, SQL dumps and log files to anyone who guessed the name. It is not any more.

Pages changed: /, /about-us, /contact, /methodology, /changelog, /api, /trending-tags, /youtube-trend, /youtube-title-generator, /youtube-tag-generator, /youtube-hashtag-generator, /youtube-tag-extractor, /youtube-tag-gap, /youtube-description-extractor, /youtube-thumbnail-downloader, /youtube-channel-id, /youtube-channel-logo-downloader, /youtube-money-calculator, /find-facebook-id, /trending-tags/india, /trending-tags/united-states, /trending-tags/pakistan, /trending-tags/indonesia, /trending-tags/bangladesh, /trending-tags/brazil, /trending-tags/egypt, /trending-tags/nigeria