Inside The Baby Mind

01The crash

This has to come before any recommendation, because it changes what kind of advice is valid at all.

The median fell 18×. April alone accounts for 61% of every view the channel has ever received. August, at a comparable publishing rate, totalled 5,340 views — less than a single average April video.

The channel is shipping roughly what it shipped at its peak and getting twenty times less for it.

02Three hypotheses, two eliminated

• ❌"Quality dropped" — disproved.

  Real frames were pulled from the videos themselves (the hq1/2/3 auto-thumbnails are frames from the video, not the cover art) and the peak video was compared against an August one.

  

  

  Production visibly improved and views fell 18×. Whatever broke, it was not the pictures.

• ❌"Thumbnails broke" — disproved.

  Peak and August covers compared side by side. The template is identical: large text on the left, illustration on the right, red arrow, red bar carrying the key phrase. Nothing changed.

• ✅"The title type changed" — supported.

  This is what actually moved.

  Title type	Videos	Median	Best
  Stakes only ("hurt", "stop now", "never")	7	3,400	450,647
  List ("10 things…", "Do these 7…")	16	3,377	604,011
  List + stakes	10	2,919	765,801
  Other	11	2,034	16,422
  "What / why" — pure curiosity	37	1,536	550,782

  And how the mix shifted month by month:

  2026-04   lists 68%    what/why 22%    -> median 7,082
  2026-05   lists 50%    what/why 14%    -> median 2,678
  2026-06   lists  6%    what/why 46%    -> median 1,294
  2026-07   lists 12%    what/why 81%    -> median 1,262
  2026-08   lists  9%    what/why 81%    -> median   381

  The channel abandoned lists at exactly the moment it started falling. Lists carry a median 2.2× higher than "what/why".

  Stated honestly: this is correlation. They may have switched format because views were already sliding, rather than the other way round. There are counter-examples — the 550,782 video is "Why Babies NEED to Sleep Next to You", a "why" title. So "what/why" can work; it simply carries half the median, and the changeover lines up with the collapse too precisely to ignore.

One more shift: video length dropped from 1,076 to 721 seconds. The peak videos were longer.

03What Tim's doctrine says — and it names this exact case

That reframes the whole teardown. A gradual monthly decline that survives a quality upgrade, holds its thumbnail template, and keeps its publishing rate is the signature of feed suppression, not of a title-format mistake. The title correlation in section 02 may well be a symptom rather than the cause — the channel changed format while sinking, which is what people do when the numbers stop working.

04The AI avatar question

This was not guessed from titles. Actual frames from the videos were examined, and they show:

• AI illustrations plus voiceover. Static drawn scenes in a cartoon style, consistent across the channel. No talking head, no avatar, no live footage.
• The channel is completely faceless. There is no presenter at all, synthetic or otherwise.
• August added burned-in subtitles over the illustrations.

What that means in practice. If the question was whether the niche is already taken by AI avatars — it is not. But the avatar approach looks like the wrong tool here regardless. The top channels in this niche are either faceless illustration plus voice, or real people: doctors, psychologists, parents.

This is a trust niche. A synthetic presenter works against you, because a parent has to believe whoever is talking about their child. Combined with section 03, the safer read is that the human contribution needs to become more visible, not less.

05Competitors

Micro-channels with extreme multiples — the most useful evidence here

These prove a reachable ceiling rather than an abstract one.

Note @timelesspoint and @DataExplainerofficial: the list format in parenting psychology, on channels of 5,000 to 12,000 subscribers, working right now. The format this channel abandoned is currently paying smaller channels.

Direct rivals, fresh — 200K+ in two months

06Adjacent niches

07What to do

1. Bring back list titles. Median 3,377 against 1,536 for "what/why". This is the only measurable difference between the peak and the collapse.
2. Bring back stakes — "hurt", "stop now", "mistake". The stakes-only format carries the channel's best median at 3,400.
3. Bring back the length. Peak videos ran 1,076 seconds, current ones 721. The drop in length coincides with the drop in views.
4. Develop the fathers angle. Already proven on this channel (450K, 114K) and outside it (@DadGoodOfficial, 85×).
5. Move into attachment and adult childhood trauma, where channels under 2,000 subscribers are pulling 153× and 813×.
6. Do not build an AI avatar. This is a trust niche; none of the channels that work use one. Faceless illustration plus voice already works and costs less.
7. Rethink the pace. Twenty-two videos a month did not save the channel when the format broke, and per section 03 that volume is itself a risk factor. Fewer and sharper beats the conveyor.

08Titles ready to shoot

1. 10 Things You Say That Hurt Your Baby's Brain (STOP NOW)
2. 7 Parenting Habits That Create Anxious Children — Most Parents Do #4
3. 8 Things Every Dad Does That Babies Remember Forever
4. Japanese Parenting Secrets That Build Confident Kids — 9 Rules
5. 6 Signs Your Baby Has a Secure Attachment (And 4 That Worry Doctors)
6. 9 Everyday Mistakes That Damage Your Toddler's Confidence

09What was verified, and what wasn't

10Sources

Other teardowns: Beauty is Tested v2 · Beauty is Tested v1