SEO Studio

Score every title against the live YouTube niche. Then rewrite it to win.

Paste a title. We pull the top 50 YouTube results for that niche, score your title on a 6-dimension rubric, surface the gap your competitors are missing, and hand back 3 AI rewrites. Plus a 300-word description and 3 hashtags pulled from real search demand. One click pushes the new title and description back to YouTube.

Score a title →See how it works

Solo plan and above · ~30 seconds per run · one-click apply via official YouTube API

Per-title scorecard

The number first. Then the rewrite that earns a higher one.

Every title gets a 0–100 score on the deterministic rubric, then up to 3 AI rewrites scored against the same rubric. The breakdown is always visible so you can see exactly which dimensions are costing you points. Keyword overlap, hook strength, length band, viral-format match.

Hybrid scoring. Deterministic SEO, AI-judged CTR + hook
Same rubric applied to your title and the AI rewrites
5-then-3 generation. Strongest titles only ever surface
Quality-floor retry if the first pass isn’t strong enough

Your title

How I doubled my YouTube views in 30 days
62/100Needs work
SEO · keyword overlap72
CTR · click-through pull58
Hook · opener strength51
Length 50–70 chars88

AI rewrite · score 91

1 of 3

I Tried 3 YouTube Strategies for 30 Days | What Truly Doubled My Views

Why: first-person opener, pipe structure, no year, anchored to the gap surfaced from competitor analysis

Six dimensions, one score

We don’t guess at what makes a title work. We measure it.

The deterministic rubric is six dimensions, each weighted to reflect how YouTube’s recommendation engine really ranks titles. Same rubric runs on your draft and on every AI rewrite, so the comparison is honest.

01

Length & character count

Your title scored against the 50–70 character sweet spot. Lengths outside that band lose CTR on mobile and get truncated on desktop search results.

02

Keyword overlap (fuzzy stem)

How many words from the niche keyword appear in the title. Using stem matching so "shop", "shopping", "shopped" all count. Anti-stuffing penalty applies above the threshold.

03

Front-loading

Whether the first three words contain a power word, question starter, or number. Front-loaded titles win the scroll because YouTube truncates after the first ~30 characters on mobile.

04

Hook structure (pipe / brackets)

Detects whether the title uses YouTube’s proven structural patterns: pipe dividers, brackets, parens. These add a second-beat curiosity layer without spending more characters.

05

Viral format match

Pattern-matches against the proven viral title formats. Listicle, transformation, contrast, journey, deep-dive. Anchors the title to a frame YouTube’s recommendation engine already understands.

06

Power words + numbers

Power-word density and presence of any number. Strict caps so the title doesn’t feel templated; the rubric rewards naturally-placed words over keyword bingo.

Before

48 words · score 38

"Hey guys! Welcome back to my channel. In this video I'm going to show you how I doubled my YouTube views. Don't forget to like and subscribe! Hit the bell icon..."

After · option 1 of 3

342 words · score 89

Visible before "Show more"

The exact 3-strategy YouTube growth experiment that doubled my channel views in 30 days. Full breakdown of what worked, what flopped, and the data behind every change.

Body excerpt

I ran three different growth tactics over 30 days. Keyword research, retention hooks, and end-screen optimization. Two of them moved the needle. One didn't. Here's the full breakdown so you can skip the dead ends...

#YouTubeGrowth#YouTubeViews#YouTubeStrategy
Description rewrites

Three descriptions. Each built for the first 150 characters.

YouTube only shows the first ~150 characters before "Show more". That’s the part that has to earn the click into your description. We rewrite from scratch: opening hook with the primary keyword, body that weaves the next 2 most important keywords once each, real CTA, and exactly 3 CamelCase hashtags pulled from autocomplete data. Not invented.

Length

300–400 words. Short ones lose search ranking.

Opening

Primary keyword in the first 150 chars. No "Welcome to my channel".

Body

Flowing paragraphs. No bullets, no sub-headers, no emoji.

Hashtags

Exactly 3. CamelCase. Derived from real autocomplete demand.

How it works

From draft title to live YouTube edit in under 60 seconds.

Five stages, all of them yours to interrupt or skip. Your title, your call. The studio just makes it the most-informed call you’ve ever made.

01

Pick the keyword

Paste your title. The studio offers 3 keyword intent options so you anchor the analysis to the right search intent before any data is fetched.

02

Live YouTube fetch

Top 50 results pulled for your niche keyword via the official Data API. Titles, view counts, channels, tags. Plus YouTube autocomplete + Serper + SerpAPI in parallel.

03

Score your title

The deterministic rubric runs on your draft. Length, keyword overlap with stem matching, front-loading, hook structure, viral format, power words.

04

AI gap + 5 rewrites

Claude Sonnet 4.6 reads the live data and your channel’s viral history, names the angle every competitor shares, the one they all miss, and writes 5 titles aimed at the gap.

05

You see the result

Score, rubric breakdown, top 3 rewrites, gap analysis, keyword opportunity table, and. If you ask for it. 3 ready-to-paste descriptions with hashtags.

Output structure

Seven distinct output blocks. Every one is publishable.

The studio doesn’t hand you a wall of text. Each block renders in its own card so you can scan, mark, and apply without re-reading. And the whole report rehydrates from the Reports tab whenever you want it back.

Title score (0–100)

Deterministic rubric across SEO, CTR, hook, and length. Same numbers every time. Plus Claude’s subjective CTR + Hook scores so subjective qualities aren’t fudged with regex.

3 AI title rewrites

Claude generates 5, the system ranks them, you see the strongest 3. Each anchored to a gap_opportunity surfaced from the live competitor data. Not generic templates.

Gap analysis from live data

The AI reads the actual top YouTube results for your niche, names the angle every competitor shares (overused), and the angle every competitor misses (your wedge).

Keyword opportunity table

Up to 15 keyword phrases scored on volume + competition. Volume from autocomplete frequency. Competition from how many top videos already target the exact phrase in their title.

3 description rewrites

300–400 words each, opening hook above the fold, primary keywords woven naturally into the body, and 3 CamelCase hashtags pulled from real autocomplete data. Not invented.

Per-video critique (vision)

On any uploaded video: title score, description verdict + rewrite, plus a Claude-vision read of the thumbnail (face, contrast, text-overlay, composition) with specific tips.

One-click apply to YouTube

Picked a rewrite? Push it back to YouTube with one click via the official Data API. We snapshot the before/after so you can track which optimizations moved views.

High-opportunity keywords

5 of 15

Phrase

Volume

Competition

Score

youtube growth strategy

HIGHLOW

92

double youtube views

MEDLOW

84

30 day youtube challenge

MEDMED

71

youtube algorithm 2026

HIGHHIGH

54

small channel tips

LOWLOW

48

Keyword opportunity

Up to 15 phrases scored on real volume + real competition.

Volume comes from how many YouTube autocomplete suggestions contain the phrase. Autocomplete only surfaces high-volume queries, so it’s a real demand signal. Competition comes from how many of the top 50 videos for your niche already target the exact phrase. Score weights low competition + decent volume. The same opportunity logic VidIQ uses, sourced from your actual niche search.

Volume signal

Autocomplete frequency. Only surfaces queries people really type.

Competition signal

Top-video title hits. Already-targeted phrases are harder to win.

By plan

How many SEO runs you get each month.

Each title analysis is one credit. Each description rewrite is one credit. Per-video critique is the description-side half of an analyze run. No double charge. Allowances are per-channel; multi-channel Agency accounts pool credits across all channels.

Free

0

SEO runs

included per month

Not included on free tier. SEO Studio is paid-only

Solo

20

SEO runs

included per month

Title + description + apply-to-YouTube · 3 channels

Most popular

Growth

50

SEO runs

included per month

Same engine, higher monthly allowance · 5 channels

Agency

150

SEO runs

included per month

Pooled across 10 channels · per-video critique included

Same hybrid rubric + Sonnet 4.6 generation across all paid plans.

See full pricing →
FAQ

The SEO engine, answered honestly.

Real answers from how the product behaves. The rubric, the rewrite logic, the apply-to-YouTube boundaries, and what it won’t do.

Still have questions? Email us →
No. The score is a hybrid: the SEO component is fully deterministic (length, keyword overlap with fuzzy stem matching, anti-stuffing checks, pipe discipline, viral format detection) so it returns the same number every time. The CTR and Hook components come from Claude’s judgement against a rubric that’s embedded in the prompt. Because regex can’t reliably score subjective qualities like emotional pull or opener strength. The combined score is what you see; the breakdown is shown so you can see exactly which dimensions cost you points.
Because most titles are ambiguous. A title like "My morning routine" could be a wellness video, a productivity video, or a parenting video. The search competition is wildly different in each case. Before the full analysis runs, the studio shows 3 keyword intent options and you pick the one that matches your actual video. That choice anchors the YouTube search, the keyword scoring, and the AI’s gap analysis. If you skip the picker, the AI will best-guess from the title alone. Still useful, just less precise.
No. Claude reads the live top 10 YouTube results for your niche keyword (titles + view counts), the YouTube autocomplete suggestions, the keyword opportunity table, and. Critically. Your channel’s own viral history. From that data it identifies the dominant pattern every competitor shares ("overused angle") and the angle nobody is using ("gap opportunity"), then writes 5 titles aimed at that gap, in your voice. The system ranks all 5 by combined score and surfaces the top 3.
Those are non-negotiable rules baked into the prompt because we tested them against thousands of titles in our beta. Colons and em-dashes feel like marketing copy and tank CTR. Year-stamped titles (2024, 2025, 2026) decay fast. A "Best apps for 2025" video stops getting clicks the day 2026 starts. The required structure (one pipe, opening beat + closing beat) is the format that consistently wins for personal vlog and tutorial content. If Claude slips on any rule the post-processor cleans it up before you see the title.
YouTube doesn’t expose true search volume publicly. We use two strong proxies. Volume: how many YouTube autocomplete suggestions contain the phrase. Autocomplete only surfaces high-volume queries, so this is a real demand signal. Competition: how many of the top 50 videos for the niche already target the exact phrase in their title plus how many tag it. The combined score weights low competition + decent volume, which is the same logic VidIQ’s "keyword opportunity" surface uses, but our pool is built from your actual niche search rather than an estimated global average.
The output is intentionally not corporate. The prompt forbids generic openers ("Welcome to my channel"), bullet lists, sub-headers, emojis, fake CTAs ("SMASH that like button"), and stock keyword stuffing. It writes 2–3 flowing paragraphs in a conversational voice with the primary keyword woven into the opening (visible before "Show more"), the next 2 most important keywords once each in the body, and exactly 3 hashtags on the final line in CamelCase format. We strip em-dashes from every output as a safety net. They’re a common AI tell.
The draft analyzer optimizes before you upload. The per-video critique runs on already-uploaded videos. It pulls the live title, current description, and thumbnail from your channel, scores all three, runs Claude vision on the thumbnail (face detection, contrast read, composition assessment, text-overlay check), and returns a rewritten description plus specific thumbnail tips. The point is to fix existing videos in your back-catalog that are underperforming relative to their topic.
It pushes the rewritten title and/or description back to YouTube via the official Data API. The same write endpoint the YouTube Studio app uses. The change is live within seconds. We snapshot the before/after view + like + comment counts at the moment of the apply so you can come back later and see whether the rewrite moved the numbers. Each apply is logged to your "Your optimizations" panel and the stats refresh lazily every 6 hours.
YouTube doesn’t penalize title changes by themselves. What hurts performance is changing a title in a way that breaks the search intent of the people who were already finding the video. E.g. swapping "How to grow basil" to "My garden tour" mid-flight. The AI rewrites are anchored to the same primary keyword as your original, so search ranking should hold or improve. We still recommend you only apply to videos where the original title is clearly underperforming. The studio will tell you when that’s the case.
Yes. Every /seo/analyze run is persisted per channel and shows up in the Reports tab. Newest first, up to 50 rows. Click any one to reopen the full analysis (score, rubric breakdown, AI rewrites, intent, gap, keyword scores, top videos for the niche). Re-running the same title updates the existing row instead of stacking duplicates, so the Reports list stays clean. Description outputs are tagged onto the most recent analysis row so the whole report rehydrates as one unit.
Title scoring works the same. The rubric isn’t format-aware because Shorts and long-form share the same ranking signals. The keyword research and competitor pool work better for long-form because Shorts thumbnails get less play in search and Shorts titles often get truncated to a few words. For Shorts-only content we recommend the title rewrites still, but pay less attention to the description critique. Shorts descriptions don’t carry meaningful SEO weight.
~25–40 seconds end-to-end. The analysis fans out: YouTube niche search (last 50 results), YouTube autocomplete + Serper related searches + SerpAPI Google autocomplete in parallel, n-gram extraction + scoring, then Claude Sonnet 4.6 for the gap analysis + 5 title rewrites. Each /seo/analyze run is one credit. The description generator is a separate one-credit charge (because it’s a separate Claude call producing 3 description options). Per-video critique is the description-side half of an analyze run. No double charge.