Ben Davis — 30-day channel report

Audience, demographics, and performance for 2026-06-24 → 2026-07-23, measured against the preceding 30 days (2026-05-25 → 2026-06-23).

@bmdavis419

41,100 subscribers

192 videos · 2,238,399 lifetime views

source: YouTube Analytics API

Views

88,811

↓ 41.3% vs prior 30d

median 1,734/day

Watch time

10,633 hrs

↓ 37.1% vs prior 30d

637,958 min

Avg view duration

7:11

↑ 7.2% vs prior 30d

28.37% of length

Net subscribers

+1,450

↓ 5.4% vs prior 30d

+1,888 / −438

Likes

2,550

↓ 45.0% vs prior 30d

28.7 per 1k views

Comments

215

↓ 59.0% vs prior 30d

2.4 per 1k views

Subs feed

43.1%

of all views

Sub retention

77%

of gains kept

Subbed vs not

8:54 / 6:25

avg duration

Top market

25.2%

United States

Core age band

68.0%

25–44 years old

Peak day

17,638

2026-07-10

Daily performance

Views and watch time move together across the window; both spike on upload days and decay over the following week.

Views per day

Vertical rules mark uploads published in this window. Hover or focus the chart for a daily readout.

Table view — daily values

Watch time per day

Estimated hours watched.

Subscribers gained vs lost

Gains above the axis, losses below; the line is the running net total.

Audience composition

Who is watching, on what, and how deeply each cohort engages.

Viewer share by age band

Percentage of views from signed-in viewers. 68.0% sit in the 25–44 range.

Viewer share by gender

Signed-in viewers only; YouTube reports gender for a subset of the audience.

Views by device

Bar length is views; the label is average view duration on that device.

Subscribed vs not subscribed

Share of views and how long each cohort stays.

Operating system by views and watch time — top 12 platforms
Platform	Views	Share	Watch hrs	Min / view
macOS	23,680	26.7%	2,574	6.5
iOS	19,511	22.0%	2,651	8.2
Windows	17,672	19.9%	1,889	6.4
Android	16,026	18.0%	1,993	7.5
Apple_Tvos	1,231	1.4%	219	10.7
Smart TV	991	1.1%	142	8.6
webOS	754	0.8%	121	9.6
Amazon_Fireos	469	0.5%	63	8.1
Rokuos	348	0.4%	52	8.9
PlayStation	112	0.1%	19	10.1
Vidaa	85	0.1%	14	9.9
Xbox	68	0.1%	9	8.1

Geography

25.2% of views come from United States. Retention varies sharply by market — worth reading alongside volume.

Views by country — top 12

Bar length is views; the trailing label is average view duration in that market.

All reported countries — top 25 by views
#	Country	Views	Share	Watch hrs	Avg duration	Avg viewed	Subs +
1	United States US	22,348	25.2%	3,125	8:23	32.2%	312
2	Germany DE	5,846	6.6%	712	7:18	28.9%	—
3	India IN	5,530	6.2%	476	5:09	21.3%	—
4	United Kingdom GB	3,781	4.3%	514	8:09	31.7%	12
5	Canada CA	3,428	3.9%	432	7:33	29.1%	—
6	Australia AU	2,582	2.9%	360	8:22	32.0%	—
7	Brazil BR	2,303	2.6%	280	7:18	29.3%	—
8	Netherlands NL	1,791	2.0%	230	7:42	30.9%	—
9	Poland PL	1,757	2.0%	201	6:51	26.6%	—
10	France FR	1,619	1.8%	180	6:40	26.2%	—
11	Indonesia ID	1,432	1.6%	126	5:16	22.2%	—
12	Spain ES	1,347	1.5%	153	6:48	27.4%	—
13	Italy IT	1,305	1.5%	135	6:11	25.6%	—
14	Sweden SE	1,141	1.3%	154	8:06	31.4%	—
15	Singapore SG	1,060	1.2%	116	6:32	26.4%	—
16	Japan JP	1,036	1.2%	116	6:42	26.8%	—
17	Russia RU	1,025	1.2%	92	5:22	22.1%	—
18	Philippines PH	964	1.1%	106	6:34	27.2%	—
19	Switzerland CH	788	0.9%	102	7:46	30.7%	—
20	Israel IL	737	0.8%	85	6:55	26.9%	—
21	Vietnam VN	708	0.8%	58	4:53	20.5%	—
22	South Africa ZA	695	0.8%	92	7:58	31.7%	—
23	Romania RO	661	0.7%	82	7:26	28.6%	—
24	Ukraine UA	654	0.7%	69	6:20	24.7%	—
25	Thailand TH	653	0.7%	70	6:25	25.4%	—

How viewers arrive

Traffic sources and playback surfaces. The subscriptions feed dominates both views and watch time, which is why upload cadence drives the daily curve so directly.

Views by traffic source

All 11 reported sources. Labels show views and share of total.

Traffic source — views vs watch time
Source	Views	Watch hrs	Min / view
Subscriptions feed	38,311	5,265	8.2
YouTube search	18,246	1,471	4.8
Direct / unknown	13,054	1,429	6.6
Suggested videos	8,615	1,295	9.0
External sites	3,564	199	3.4
Other YouTube pages	2,790	386	8.3
Channel page	2,422	282	7.0
Playlists	1,124	220	11.8
Notifications	451	62	8.3
End screens	218	22	6.2
Shorts feed	4	0	0.0

Playback location and share destinations
Playback surface	Views	Share
Watch page	83,515	94.0%
Browse / feed	4,195	4.7%
Embedded player	1,008	1.1%
Shorts feed	80	0.1%
Channel page	1	0.0%
Share destination	Shares	Share
Copy link	441	63.8%
Other	167	24.2%
WhatsApp	33	4.8%
Email	10	1.4%
System share sheet	10	1.4%
Text message	8	1.2%
WhatsApp Business	7	1.0%
Gmail	6	0.9%
Discord	6	0.9%
Facebook	3	0.4%

Video performance

3 of the top 20 videos were published inside the window; the rest are back catalogue still earning views. Retention is the clearer signal here — view counts favour recency.

Views vs percentage watched — top 20 videos

Horizontal position is views, vertical is average percentage of the video watched. Dot size is watch time; filled dots were published in this window.

Top 20 videos by views in window · ↑ marks an upload inside the window
#	Video	Views	Watch hrs	Avg duration	Avg viewed	Subs +	Likes	Comments
1	GPT-5.6 is better than Fable ↑ 2026-07-10 · 29:57	23,058	3,735	9:43	32.5%	248	894	83
2	How I Turned Pi Into the Ultimate Coding Agent 2026-05-15 · 21:27	20,667	1,809	5:15	24.5%	148	299	10
3	You're probably using GPT-5.6 wrong, here's how to fix it ↑ 2026-07-16 · 27:50	15,080	2,339	9:18	33.4%	167	614	52
4	I Created the Ultimate Coding Agent by Combining Pi, Codex, and Claude Code... ↑ 2026-07-24 · 30:35	7,258	996	8:14	26.9%	90	367	43
5	Can Open Source Models Compete with the Frontier? 2026-06-23 · 22:58	3,297	371	6:44	29.4%	25	44	7
6	My New Favorite Agent 2026-06-16 · 24:41	1,973	206	6:15	25.4%	22	50	2
7	My Mac Sucked, Here's How I Fixed It... 2025-10-08 · 19:53	1,604	106	3:58	20.0%	14	32	1
8	Why is Everyone Switching to Svelte? 2026-03-19 · 23:03	1,151	92	4:48	20.9%	10	36	1
9	Pi: The Best Coding Agent You've (Probably) Never Heard Of 2026-04-01 · 22:03	1,084	70	3:51	17.5%	8	22	0
10	What's the Best Way to Deploy Your Apps? 2025-11-07 · 21:18	920	75	4:54	23.0%	6	14	0
11	I Hated Monorepos, Now I Can't Build Without Them 2026-02-01 · 16:18	890	57	3:50	23.6%	5	18	0
12	This is THE Framework You Should be Using in 2026 2026-01-06 · 20:33	798	52	3:55	19.1%	10	18	2
13	I Can't Believe This Actually Works 2026-01-16 · 20:45	735	34	2:46	13.4%	6	1	0
14	I Replaced the Project I Spent Months on With a Markdown File 2026-05-19 · 23:57	733	71	5:49	24.3%	8	18	1
15	CRON Jobs Explained and an Example put in Production in 5 minutes! 2022-11-17 · 5:30	681	26	2:19	42.4%	2	9	1
16	The thing that makes Svelte 5 special 2025-05-02 · 10:40	522	32	3:41	34.6%	4	7	2
17	I Was Wrong About Svelte... 2025-10-12 · 9:00	450	20	2:39	29.6%	0	10	0
18	I was wrong about GPT 5.5 2026-05-28 · 17:34	428	25	3:32	20.1%	2	7	0
19	Functional Programming is Amazing (elixir is my new favorite language) 2025-02-03 · 8:49	425	20	2:50	32.2%	3	15	1
20	You Should Try Solid 2026-01-19 · 31:03	409	44	6:31	21.0%	3	8	0

Period comparison

Every metric available for both windows, indexed so direction and magnitude are directly comparable.

Change vs prior 30 days

Percentage change, current window against the preceding 30 days. Zero line is the prior period.

Metric detail — current vs prior window
Metric	Current 2026-06-24 → 2026-07-23	Prior 2026-05-25 → 2026-06-23	Change
Views	88,811	151,195	↓ 41.3%
Watch time (hours)	10,633	16,892	↓ 37.1%
Average view duration	7:11	6:42	↑ 7.2%
Average percentage viewed	28.37%	30.54%	↓ 7.1%
Subscribers gained	1,888	2,147	↓ 12.1%
Subscribers lost	438	614	↓ 28.7%
Net subscribers	+1,450	+1,533	↓ 5.4%
Likes	2,550	4,635	↓ 45.0%
Comments	215	524	↓ 59.0%
Shares	702	1,094	↓ 35.8%

Notes on reading this

Dislikes are reported for the current window only (90, a 28:1 like ratio) and are excluded from the comparison table because the prior-window figure was not returned by the API.

Demographic percentages cover signed-in viewers only, so they describe the measurable audience rather than everyone who watched. Peak day was 2026-07-10 at 17,638 views; the quietest was 2026-07-09 at 1,344.