1
00:00:00,060 --> 00:00:08,810
OK in a previous video we talked about thumbnail A B testing a meaning in advanced analytics lecture.

2
00:00:08,970 --> 00:00:14,700
And today I'm going to I wanted to go a bit deeper into it to just tell you and show you how you can

3
00:00:14,700 --> 00:00:21,700
actually do advanced a bit thumbnail of A B testing and what it is useful for.

4
00:00:22,320 --> 00:00:25,070
So there are actually two ways to do a thumbnail A B testing.

5
00:00:25,080 --> 00:00:27,610
One is the paid way and one is a free way.

6
00:00:27,690 --> 00:00:31,090
So the paid way is to do it with the two buddy.

7
00:00:31,100 --> 00:00:33,480
This is the two buddy application.

8
00:00:33,540 --> 00:00:40,250
You can go to your dashboard and you can go and do the two buddy A B testing from this one.

9
00:00:40,290 --> 00:00:47,220
I'm not gonna show you this right now because it is the paid option and it requires the legend subscription

10
00:00:47,220 --> 00:00:51,060
from two buddy legend is the most expensive subscription for the two.

11
00:00:51,070 --> 00:00:58,010
But let me just go to two buddy dot com and show you what the legend subscription actually is.

12
00:00:58,050 --> 00:01:00,600
I'm going to show you the pricing right here and there.

13
00:01:00,630 --> 00:01:06,000
This is the most expensive forty nine dollars which is a very expensive.

14
00:01:06,000 --> 00:01:09,640
And this is the only way you can do it internally.

15
00:01:09,660 --> 00:01:14,070
What it does basically what the AP testing does is you upload it to different thumbnails or you can

16
00:01:14,070 --> 00:01:17,480
upload two different titles or even descriptions or even tax ads.

17
00:01:17,580 --> 00:01:25,230
And it is rotating every single day or two for it and in 24 hour basis and measuring the a b the results

18
00:01:25,650 --> 00:01:27,430
every single day.

19
00:01:27,480 --> 00:01:34,440
This is the easiest way and the best way to do it because it does not require a lot of working from

20
00:01:34,440 --> 00:01:39,490
you and overall after a few days it's gonna give you results whether the thumbnail is better or not.

21
00:01:39,720 --> 00:01:40,830
That's a very good way.

22
00:01:40,830 --> 00:01:45,460
But for the beginners or for the channels that are not generating enough income.

23
00:01:45,750 --> 00:01:52,500
This is a bit steep and and a bit expensive so other way is to do it yourself.

24
00:01:52,560 --> 00:01:59,600
So how you can do a b testing is you can create a two different sets of thumbnails.

25
00:01:59,610 --> 00:02:03,670
You can create two different sets of thumbnails for each and every video.

26
00:02:03,780 --> 00:02:09,330
If you want to if this is your first video of this the first time you upload this type of video you

27
00:02:09,330 --> 00:02:12,320
can create two different thumbnails.

28
00:02:12,330 --> 00:02:18,910
And what you can do is you can go and test your thumbnails for three days.

29
00:02:18,930 --> 00:02:23,970
But remember for example if you take the video of test of the thumbnail for three days go for total

30
00:02:23,990 --> 00:02:27,800
go for custom and tested for let's say if you are.

31
00:02:28,140 --> 00:02:33,540
Let's give as an example f a loaded if you upload it is on April 1st test it for me prefers to April

32
00:02:33,540 --> 00:02:37,850
3rd and see what is your city are like overall.

33
00:02:37,950 --> 00:02:47,340
I do not suggest you to go with just three days because it might change a bit and then what you will

34
00:02:47,340 --> 00:02:50,040
do I suggest that you go for a week at least.

35
00:02:50,040 --> 00:02:54,660
But if you can go for the four this one as well it's not a problem for three days as well.

36
00:02:55,110 --> 00:03:01,260
And then after this change your thumbnail and test it for another three days or another week depending

37
00:03:01,260 --> 00:03:09,530
on what time you do it and whatever you see the differences see in a custom in the always check over

38
00:03:09,530 --> 00:03:13,960
the custom dates like for example if you need a form from 4th to 6.

39
00:03:14,000 --> 00:03:19,670
Check check from 4 to six and ignore whatever it is as you can see in this case the CCR went down even

40
00:03:19,670 --> 00:03:22,580
though I didn't change the thumbnail so it is still the same.

41
00:03:22,580 --> 00:03:33,230
And then do it two or three times do you will see approximately which thumbnail is working better approximately.

42
00:03:33,230 --> 00:03:39,950
This is not going to be exact thing and you see if the city are from one thumbnail is higher than from

43
00:03:39,950 --> 00:03:46,390
another thumbnail then that means you understand that one thumbnail is actually working better already.

44
00:03:46,400 --> 00:03:52,520
You see I don't think that you can do is whatever you upload the thumbnail and you see the Saeki are

45
00:03:52,960 --> 00:03:56,120
of it lower than your channel's average.

46
00:03:56,120 --> 00:03:58,470
If it's a lower then let's say this one.

47
00:03:58,490 --> 00:04:03,290
Your channels average for this amount of data or an hour the one week for example if you upload it is

48
00:04:03,290 --> 00:04:07,230
for one week and you see it's actually lower than the average.

49
00:04:07,400 --> 00:04:14,570
Only after that work on for the same amount of time due the other thumbnail if you see the thumbnail

50
00:04:14,570 --> 00:04:21,820
is still not doing the greatest for next week do the third one and etc etc. until you find out the best

51
00:04:21,820 --> 00:04:28,520
and all that works for you because you need to work on city are 100 percent CAGR is the thing that drives

52
00:04:28,520 --> 00:04:34,730
the views for you when you have the high since ya YouTube will try it will start pushing your videos

53
00:04:34,730 --> 00:04:41,210
more and impressions will increase because what YouTube is trying to do is trying to decrease your city

54
00:04:41,210 --> 00:04:42,530
are as high as possible.

55
00:04:42,830 --> 00:04:43,960
What does it mean.

56
00:04:43,970 --> 00:04:49,940
It means that it's tried to try to show you do your video to more and more people and if the more people

57
00:04:49,940 --> 00:04:54,890
chose your video the less people actually click on the certain parts percentage will click to your videos

58
00:04:55,160 --> 00:04:59,640
and they wanted to just get your videos to the certain type of cities.

59
00:04:59,640 --> 00:05:06,830
For example if YouTube would show my video 2 million random people and generate a million impressions

60
00:05:07,270 --> 00:05:09,490
much each year I would immediately Huff.

61
00:05:09,560 --> 00:05:14,360
It will not be 7 percent because the random people might not watch the my videos.

62
00:05:14,360 --> 00:05:15,080
They might not.

63
00:05:15,110 --> 00:05:16,670
Or they might or they might.

64
00:05:16,670 --> 00:05:21,350
Most likely they will not and it will just drop my city are YouTube does not want your city art to be

65
00:05:21,350 --> 00:05:22,380
very very low.

66
00:05:22,400 --> 00:05:28,390
They are aiming for the certain city for this and for the channels and they are trying to do it.

67
00:05:28,400 --> 00:05:29,600
Spread your channel.

68
00:05:29,600 --> 00:05:37,430
Depending on your city ya so work on your thumbnail A B testing do different thumbnails.

69
00:05:37,460 --> 00:05:41,810
I cannot tell exactly what thumbnails will work for you because it is completely different for each

70
00:05:41,810 --> 00:05:43,100
and every individual channel.

71
00:05:43,110 --> 00:05:49,790
First of all despite this type of dance being the best on the channel doing the doing overall generating

72
00:05:50,270 --> 00:05:57,620
the most watched time this video was generating the most star because this thumbnail has been here the

73
00:05:57,620 --> 00:06:04,820
longest then a longer than this once and youtube no so a type of people are watching this and that's

74
00:06:04,820 --> 00:06:08,800
why the impressions are going that high that quick as you can see it.

75
00:06:08,810 --> 00:06:14,300
It's even the speed is even increasing on the later dates because it gets it that growth is a bit gradual

76
00:06:14,300 --> 00:06:17,240
right here and here is a bit even faster.

77
00:06:17,240 --> 00:06:26,180
So work on A B testing it is not necessary again it is more of an advanced stuff if you have everything

78
00:06:26,660 --> 00:06:28,670
other than every other thing.

79
00:06:28,940 --> 00:06:34,220
Very well if you have high average watch them in everything and you want higher since ya this is the

80
00:06:34,220 --> 00:06:36,750
way to grow your CTA.

81
00:06:36,770 --> 00:06:38,540
So this will be it for this one let's go to the next one.
