1
00:00:00,000 --> 00:00:08,720
So, hello and welcome everybody.

2
00:00:08,720 --> 00:00:14,280
You might today have heard already a lot about Matomo, from different angles, from different

3
00:00:14,280 --> 00:00:22,160
use cases, from beginner-friendly content to topics that are focusing quite on experts.

4
00:00:22,160 --> 00:00:29,920
And I guess a lot of questions about the topic of Matomo, beginning from usability to technical

5
00:00:29,920 --> 00:00:33,000
questions, a lot of questions might have arised.

6
00:00:33,000 --> 00:00:38,440
So that's why today we have a Q&A session, where you can ask these questions you have

7
00:00:38,440 --> 00:00:43,440
about Matomo to these speakers we have here today.

8
00:00:43,440 --> 00:00:49,060
So the way this works is quite similar to the talks we have already had today.

9
00:00:49,060 --> 00:00:52,480
You can see obviously the live streamer if you're watching.

10
00:00:52,480 --> 00:00:58,240
And we have a chat room for this where you can ask any question you have about Matomo.

11
00:00:58,240 --> 00:01:04,680
And I will then forward this question to one of the five speakers we have here in the room.

12
00:01:04,680 --> 00:01:10,560
So to give you a bit of time to collect your questions and post them in the chat room,

13
00:01:10,560 --> 00:01:15,440
I can start by quickly introducing our speakers here.

14
00:01:15,440 --> 00:01:17,320
So let's start with Ronan.

15
00:01:17,320 --> 00:01:25,080
Ronan is a Matomo analytics expert since 2010, and he's mostly focusing on training and doing

16
00:01:25,080 --> 00:01:32,000
consulting for all types of companies, institutions, always around about Matomo analytics.

17
00:01:32,000 --> 00:01:40,500
Stefan Koning, our second expert here, is also an expert mostly on focusing on digital analytics

18
00:01:40,500 --> 00:01:42,960
and conversion rate optimization.

19
00:01:42,960 --> 00:01:48,080
He has a lot of experience with Matomo and he will be able to answer your analytics related

20
00:01:48,080 --> 00:01:49,080
questions.

21
00:01:49,080 --> 00:01:55,680
Then we have Thomas Zernig, he's a CEO and web analytics expert from the border between

22
00:01:55,680 --> 00:01:57,520
Austria and Germany.

23
00:01:57,520 --> 00:02:03,120
And you might know him from the German Matomo explanation videos he does, or from the Power

24
00:02:03,120 --> 00:02:08,080
of the Extension Matomo helper, which might some of you will be using.

25
00:02:08,080 --> 00:02:14,600
And he will be able to answer your question about how to use Matomo and how to solve issues

26
00:02:14,600 --> 00:02:17,040
using the user interface.

27
00:02:17,040 --> 00:02:21,800
And we also have here as our fourth speaker, Stefan Giel.

28
00:02:21,800 --> 00:02:25,480
He is part of the Matomo team since 2010.

29
00:02:25,480 --> 00:02:30,480
And he is working full time on the Matomo core development in the last years.

30
00:02:30,480 --> 00:02:36,520
So he is part of the core development team and expert for all the code related questions

31
00:02:36,520 --> 00:02:38,560
in Matomo.

32
00:02:38,560 --> 00:02:44,160
And besides this, he is also maintaining the Matomo translation and device detector, the

33
00:02:44,160 --> 00:02:51,720
separate project, which converts user agents into the nice device information reports we

34
00:02:51,720 --> 00:02:53,720
can see in Matomo.

35
00:02:53,720 --> 00:02:56,320
And lastly, we also have Peter Bolleke.

36
00:02:56,320 --> 00:03:01,680
He's part of the Matomo community from the very beginning in 2008.

37
00:03:01,680 --> 00:03:06,800
And since then, he's supporting companies with all operations around Matomo.

38
00:03:06,800 --> 00:03:15,440
And he can answer all kinds of technical questions around Matomo.

39
00:03:15,440 --> 00:03:25,360
And yeah, I hope everyone can see the screen.

40
00:03:25,360 --> 00:03:30,160
And let's just wait a few minutes for questions.

41
00:03:30,160 --> 00:03:36,560
Maybe in the meantime, while people are typing questions, some of you, if anyone of you wants

42
00:03:36,560 --> 00:03:48,400
to do a quick introduction or talk about some topic, they can talk about it now.

43
00:03:48,400 --> 00:03:52,640
Maybe a question to all.

44
00:03:52,640 --> 00:03:58,240
How do you learn something about Matomo or what is your entry point or what was your

45
00:03:58,240 --> 00:03:59,800
entry point?

46
00:03:59,800 --> 00:04:11,680
Or where do you can ask questions beyond this Matomo camp?

47
00:04:11,680 --> 00:04:19,200
First of all, maybe not everyone knows that there is a forum, a board, where you can ask

48
00:04:19,200 --> 00:04:26,280
questions, where you can ask questions in mother tongue, like Germany.

49
00:04:26,280 --> 00:04:30,120
There is a good support in there.

50
00:04:30,120 --> 00:04:38,920
And the other one is the Matomo support if you take the Matomo Cloud version.

51
00:04:38,920 --> 00:04:48,000
And maybe the other ones have also some tips how to get more knowledge about Matomo.

52
00:04:48,000 --> 00:04:49,000
A good point.

53
00:04:49,000 --> 00:04:52,320
I forgot to introduce myself, of course.

54
00:04:52,320 --> 00:04:58,520
I'm Lucas Winkler, and most of you will know me from the Matomo forum.

55
00:04:58,520 --> 00:05:04,840
I spend a lot of my time answering people's questions on the Matomo forum, but also maintaining

56
00:05:04,840 --> 00:05:09,320
some other documentation and answering to GitHub issues.

57
00:05:09,320 --> 00:05:13,640
And this is also kind of the answer to your question.

58
00:05:13,640 --> 00:05:19,160
If you have any questions about Matomo, if it's a general question, something doesn't

59
00:05:19,160 --> 00:05:25,320
work on your setup, something like this, the best way probably is the Matomo forum.

60
00:05:25,320 --> 00:05:30,760
A lot of people already have asked their questions there, and it's a great way to slowly get

61
00:05:30,760 --> 00:05:33,360
to the bottom of the issue.

62
00:05:33,360 --> 00:05:38,920
If you know that the thing you're asking is definitively a bug, and you have definitive

63
00:05:38,920 --> 00:05:45,000
steps how to reproduce them, you can skip the forum and go straight to GitHub and create

64
00:05:45,000 --> 00:05:50,080
a bug issue.

65
00:05:50,080 --> 00:05:57,440
And that way, ideally describe it in a lot of detail so people can reproduce it.

66
00:05:57,440 --> 00:06:04,720
And this way, people can see it, it's quickly organized, and it's easier to fix them.

67
00:06:04,720 --> 00:06:05,720
Okay.

68
00:06:05,720 --> 00:06:09,920
While we did this introduction, there have been a lot of questions.

69
00:06:09,920 --> 00:06:15,120
And I'll start with one question from Peter Jones.

70
00:06:15,120 --> 00:06:19,560
He asks, we can add the users in the Matomo interface.

71
00:06:19,560 --> 00:06:24,600
I wonder if it will be possible to use groups in the future.

72
00:06:24,600 --> 00:06:29,480
This will be useful when there are hundreds of thousands of users.

73
00:06:29,480 --> 00:06:35,400
Does anyone have to say something about this?

74
00:06:35,400 --> 00:06:36,920
Maybe it's Stefan.

75
00:06:36,920 --> 00:06:42,480
I personally don't know about this suggestion.

76
00:06:42,480 --> 00:06:48,840
I'm not even sure if someone has even suggested it until now.

77
00:06:48,840 --> 00:06:53,840
But indeed, there are Matomo instances with a large amount of users, and some kind of

78
00:06:53,840 --> 00:07:01,280
grouping might be useful, especially to give access rights to a site to many people at

79
00:07:01,280 --> 00:07:02,280
once.

80
00:07:02,280 --> 00:07:12,440
There had been a couple of requests for a more extensive permission management in Matomo.

81
00:07:12,440 --> 00:07:18,960
And yeah, some of them were, I guess, asking for groups, but there currently aren't any

82
00:07:18,960 --> 00:07:20,960
plans.

83
00:07:20,960 --> 00:07:31,680
Then again, as a workaround until then, all of the user management features can be also

84
00:07:31,680 --> 00:07:33,560
accessed via the API.

85
00:07:33,560 --> 00:07:40,560
So if the issue is just having to manually add hundreds of users to a site, then of course,

86
00:07:40,560 --> 00:07:45,800
you can easily automate this with a script in the programming language of your choice

87
00:07:45,800 --> 00:07:52,600
and add the users there.

88
00:07:52,600 --> 00:07:54,320
I hope this roughly answers your question.

89
00:07:54,320 --> 00:07:57,240
There's a bit of a delay, so I will go on.

90
00:07:57,240 --> 00:08:02,480
And if you have any follow up to this, simply ask again.

91
00:08:02,480 --> 00:08:08,120
So Valentin asks, which killer feature from Google Analytics do you think should be included

92
00:08:08,120 --> 00:08:13,040
in Matomo?

93
00:08:13,040 --> 00:08:18,040
Secondary dimensions.

94
00:08:18,040 --> 00:08:27,920
There is the custom report, it's a very great plugin for Matomo Cloud, or you can pay for

95
00:08:27,920 --> 00:08:30,480
the on-premise version.

96
00:08:30,480 --> 00:08:42,760
But I missed something like secondary dimension in time to filter in time, not to make the

97
00:08:42,760 --> 00:08:50,720
custom report and then maybe the data takes a little bit longer to process.

98
00:08:50,720 --> 00:09:00,560
So I wish, I think I added a feature request to the second dimension feature, but on the

99
00:09:00,560 --> 00:09:08,800
technical side, it's a very huge feature and I think it's very complicated to integrate

100
00:09:08,800 --> 00:09:09,800
into Matomo.

101
00:09:09,800 --> 00:09:22,320
I think the data visualization part is as well missing somehow, highly asked by a Google

102
00:09:22,320 --> 00:09:25,040
Analytics user who was Google Data Studio.

103
00:09:25,040 --> 00:09:32,720
You will have a conference tomorrow from Baptiste Tivand about how to use Metabase with Matomo

104
00:09:32,720 --> 00:09:38,720
Analytics, but it's still something which is kind of tacky, so you cannot put it within

105
00:09:38,720 --> 00:09:40,560
the end of a marketer.

106
00:09:40,560 --> 00:09:47,240
So I think working more on the integration of Matomo with a third party data visualization

107
00:09:47,240 --> 00:10:02,040
system or to have its own more sexy data visualization system is of paramount importance for Matomo.

108
00:10:02,040 --> 00:10:08,480
One feature that could be included that Valentin suggested that, now that I read it, also find

109
00:10:08,480 --> 00:10:13,920
a good idea is simply a live view that shows the number of hits in the last five minutes

110
00:10:13,920 --> 00:10:21,680
to get a really close how many people are accessing the website right now view, which

111
00:10:21,680 --> 00:10:26,640
incidentally, now that I'm looking at the reports for the Matomo camp, might also be

112
00:10:26,640 --> 00:10:32,240
something interesting to see.

113
00:10:32,240 --> 00:10:38,920
In addition to I agree with Thomas as well, definitely secondary dimensions, but also

114
00:10:38,920 --> 00:10:44,960
a decent way to import and add additional data into Matomo to combine it with the data

115
00:10:44,960 --> 00:10:52,640
set that you already have in the thing that would make a lot of sense as well.

116
00:10:52,640 --> 00:11:01,120
Okay, I see an interesting another question from Peter Jones.

117
00:11:01,120 --> 00:11:06,600
They have an older PVIC instance, but are trying to migrate it to Matomo.

118
00:11:06,600 --> 00:11:10,800
What are your experiences of migrating PVIC data into Matomo?

119
00:11:10,800 --> 00:11:15,120
Maybe something for Peter or Stefan.

120
00:11:15,120 --> 00:11:20,920
Well, I can take on this one.

121
00:11:20,920 --> 00:11:21,920
It is possible.

122
00:11:21,920 --> 00:11:26,120
It depends on how much data do you have in the instance.

123
00:11:26,120 --> 00:11:31,320
So also with low amounts of data and also with high amounts of data, it's definitely

124
00:11:31,320 --> 00:11:32,320
possible.

125
00:11:32,320 --> 00:11:37,800
But it can with a higher amount of data take some time because a lot of scheme changes

126
00:11:37,800 --> 00:11:41,000
for the MySQL database are necessary.

127
00:11:41,000 --> 00:11:50,720
And also things can break because the version jump from two or three to four is a major

128
00:11:50,720 --> 00:11:51,720
one.

129
00:11:51,720 --> 00:11:57,280
So a lot of things in the database scheme have been optimized and have been changed

130
00:11:57,280 --> 00:11:58,840
also.

131
00:11:58,840 --> 00:12:04,080
So you should definitely make sure to have a real good backup before you start doing

132
00:12:04,080 --> 00:12:06,080
that migration.

133
00:12:06,080 --> 00:12:14,160
And then think also about some outage of your Matomo instance while you're updating, at

134
00:12:14,160 --> 00:12:18,800
least with a bigger amount of data you have in your instance.

135
00:12:18,800 --> 00:12:25,460
Lock around could be, for example, like you could just leave the tracker running or writing

136
00:12:25,460 --> 00:12:27,160
into a log file.

137
00:12:27,160 --> 00:12:33,840
And then after your migration is complete, you can import this log file with the replay

138
00:12:33,840 --> 00:12:36,540
log functionality of the log importer.

139
00:12:36,540 --> 00:12:39,780
So you don't lose too much data.

140
00:12:39,780 --> 00:12:46,240
So the replay logging probably is the best solution for importing nearly this data that

141
00:12:46,240 --> 00:12:51,160
would have come into the instance while you had this outage.

142
00:12:51,160 --> 00:12:58,240
But all in all, you have to make sure that you have a few hours of outage in your Matomo

143
00:12:58,240 --> 00:12:59,240
instance.

144
00:12:59,240 --> 00:13:03,280
So your users have to know and the stakeholders have to know.

145
00:13:03,280 --> 00:13:11,320
But all in all, I also updated, I think, 20 to 30 big instances to Matomo from Pivik or

146
00:13:11,320 --> 00:13:18,320
from the early versions, and everything went perfectly well.

147
00:13:18,320 --> 00:13:24,200
One minor thing I also want to add is something a lot of, even if it's a small instance, something

148
00:13:24,200 --> 00:13:30,460
a lot of people are struggling with, if they have a really old Pivik version, is that this

149
00:13:30,460 --> 00:13:35,800
really old version, of course, also runs with a really old version of PHP, and there might

150
00:13:35,800 --> 00:13:40,680
not even be a version of PHP that is really old Pivik version is compatible with.

151
00:13:40,680 --> 00:13:43,880
And the newest Matomo version is compatible with.

152
00:13:43,880 --> 00:13:49,760
So if it's a really old version, it might be useful to do a few intermediate jumps of

153
00:13:49,760 --> 00:13:52,520
a few intermediate versions of Matomo.

154
00:13:52,520 --> 00:13:57,760
And maybe if the compatibility requirements also require it, also intermediate versions

155
00:13:57,760 --> 00:14:03,640
of PHP to make sure the migration runs as smooth as possible.

156
00:14:03,640 --> 00:14:09,560
A lot of people might not know it, but on builds.matomo.org, the download page, you can

157
00:14:09,560 --> 00:14:17,320
download every single Matomo version, I think that has been released since 2008.

158
00:14:17,320 --> 00:14:31,240
So this way you can easily update to another old version before you do the next jump.

159
00:14:31,240 --> 00:14:38,080
Another question from Juliet is, is there a possibility to aggregate that data directly

160
00:14:38,080 --> 00:14:39,240
in Matomo?

161
00:14:39,240 --> 00:14:44,400
So the number of users grouped by a certain custom variable, for instance.

162
00:14:44,400 --> 00:14:50,760
She says I'm mostly using the API and then aggregate that data with Python pandas.

163
00:14:50,760 --> 00:14:56,480
Personally, I can't say much helpful things about it.

164
00:14:56,480 --> 00:15:01,680
Don't really have much experience with any use case where you would have to aggregate

165
00:15:01,680 --> 00:15:02,920
data.

166
00:15:02,920 --> 00:15:08,960
Of course, the way you are doing it, using the API and then doing any data processing

167
00:15:08,960 --> 00:15:15,760
afterwards, of course, always works, but might not be the most ideal one.

168
00:15:15,760 --> 00:15:20,040
Probably Juliet, it would be good to elaborate a bit on that, what you're trying to achieve,

169
00:15:20,040 --> 00:15:28,280
because in the custom dimension reports, you have everything grouped by the custom dimension

170
00:15:28,280 --> 00:15:34,120
in the sense of that every custom dimension has its own row in that report.

171
00:15:34,120 --> 00:15:41,040
If you're trying to group all other reports in the dashboard, for example, for those custom

172
00:15:41,040 --> 00:15:46,480
dimensions, combinations or a single custom dimension, this is probably not possible in

173
00:15:46,480 --> 00:15:48,040
the dashboard at least.

174
00:15:48,040 --> 00:15:57,600
So going with an external solution like Python pandas or Tableau Power BI Metabase or something

175
00:15:57,600 --> 00:16:02,840
like that would be the way to go, at least for the moment.

176
00:16:02,840 --> 00:16:19,800
One quick note to Peter's previous explanation for replaying the logs by Chris is if one

177
00:16:19,800 --> 00:16:27,120
tries to go this way and jump the downtime of the upgrade by afterwards replaying the

178
00:16:27,120 --> 00:16:31,880
server logs that reach the Matomo server, one of course has to pay attention that all

179
00:16:31,880 --> 00:16:37,320
requests sent to Matomo are GET requests, not POST requests, otherwise, of course, the

180
00:16:37,320 --> 00:16:45,560
data won't be written in the reports in the server logs.

181
00:16:45,560 --> 00:16:49,320
Mikael has another question about database speed.

182
00:16:49,320 --> 00:16:55,880
He says that the database speed is a big issue and is it possible or has someone tried to

183
00:16:55,880 --> 00:17:00,960
use Solr or any other database for quicker queries?

184
00:17:00,960 --> 00:17:09,120
Maybe Stefan or I can say something about it.

185
00:17:09,120 --> 00:17:17,200
As far as I know, there aren't any plans to support other databases because it's hard

186
00:17:17,200 --> 00:17:23,240
to start supporting other databases when you have a software that should be available for

187
00:17:23,240 --> 00:17:32,520
every user on a hosted system and MySQL is simply the mostly supported database.

188
00:17:32,520 --> 00:17:40,320
Indeed, the main issue, I think, is if you wanted to add support for another database

189
00:17:40,320 --> 00:17:47,440
and it doesn't matter if it's something like Postgres SQL or something completely different,

190
00:17:47,440 --> 00:17:53,920
it would always mean you'd have to rewrite a huge portion of Matomo.

191
00:17:53,920 --> 00:17:59,480
And in this case, it might even be quicker to simply export the data from Matomo.

192
00:17:59,480 --> 00:18:07,920
And if you're already rewriting it from scratch, do the queries in this database directly.

193
00:18:07,920 --> 00:18:13,320
And of course, the other thing is if you would rewrite a huge portion of Matomo to support

194
00:18:13,320 --> 00:18:21,440
other databases that are maybe more specialized for analytics data, is that either both versions

195
00:18:21,440 --> 00:18:27,040
that would differ quite a lot, usually would need to be maintained in the future, which

196
00:18:27,040 --> 00:18:34,400
is really hard to do, or the MySQL, only MySQL compatible version would have to be dropped.

197
00:18:34,400 --> 00:18:39,200
But this would mean that nearly everyone who can at the moment simply install Matomo on

198
00:18:39,200 --> 00:18:45,720
their server using MySQL, MariaDB or something like this, would have major issues continuing

199
00:18:45,720 --> 00:18:48,440
to use Matomo in the future.

200
00:18:48,440 --> 00:18:56,360
So because of all of this, I think it won't change in the near future, that Matomo will

201
00:18:56,360 --> 00:19:00,760
be MySQL and MariaDB only.

202
00:19:00,760 --> 00:19:08,800
An interesting question by Peter Makers is, are there any ideas about integrating a website

203
00:19:08,800 --> 00:19:14,200
carbon calculator, which measures how much impact your website has on the planet?

204
00:19:14,200 --> 00:19:19,520
It could be a unique selling point for potential new users.

205
00:19:19,520 --> 00:19:25,720
If no one else has something to say about this.

206
00:19:25,720 --> 00:19:31,480
I think this is one of the use cases, which would be a really nice Matomo plugin, because

207
00:19:31,480 --> 00:19:36,240
it's something only a few people would be interested in, but it would be a nice feature

208
00:19:36,240 --> 00:19:43,760
for them, and it's something you can quite easily put around a Matomo without having

209
00:19:43,760 --> 00:19:47,720
to modify something of the core code.

210
00:19:47,720 --> 00:19:51,680
So this would be a nice, simple plugin.

211
00:19:51,680 --> 00:19:56,720
And if someone gets around to programming it, it would be really nice if they could

212
00:19:56,720 --> 00:20:01,360
publish it on the marketplace, I think.

213
00:20:01,360 --> 00:20:08,240
In the meantime, maybe a question about translation.

214
00:20:08,240 --> 00:20:13,760
How does the translation process work in Matomo?

215
00:20:13,760 --> 00:20:21,680
If Matomo isn't completely translated in the language I speak, or if I notice a type or

216
00:20:21,680 --> 00:20:27,000
a sentence that doesn't really make sense in Matomo, how can I fix this?

217
00:20:27,000 --> 00:20:28,800
How can I report this?

218
00:20:28,800 --> 00:20:34,520
Maybe Stefan has something to say about this.

219
00:20:34,520 --> 00:20:41,840
We recently switched our translation platform to Contribute2, and all our translations are

220
00:20:41,840 --> 00:20:44,920
currently hosted on Weblate.

221
00:20:44,920 --> 00:20:49,240
It's another open source software to maintain translations.

222
00:20:49,240 --> 00:20:56,600
And everyone can suggest or change the translations there to any language.

223
00:20:56,600 --> 00:21:06,240
And from there, the translations are automatically pushed to our GitHub repository and will later

224
00:21:06,240 --> 00:21:16,200
be included in any new release of Matomo or any of our plugins.

225
00:21:16,200 --> 00:21:33,600
OK, then a bit more analytics-based question from Arribas is, they are currently using

226
00:21:33,600 --> 00:21:42,160
Matomo in order to save page data, but would we like to be able to categorize the pages

227
00:21:42,160 --> 00:21:43,960
based on the department?

228
00:21:43,960 --> 00:21:46,560
The data for this is already in the page.

229
00:21:46,560 --> 00:21:52,440
And could you please let us know how to proceed?

230
00:21:52,440 --> 00:21:54,080
I can take this one.

231
00:21:54,080 --> 00:21:59,400
It's basically what you could do is, if the information is already there, you should be

232
00:21:59,400 --> 00:22:05,440
able to store those kind of input within custom dimensions and then basically also be able

233
00:22:05,440 --> 00:22:08,440
to segment on this later on as well in Matomo.

234
00:22:08,440 --> 00:22:15,840
I would recommend to start measuring it together with the default page view tracking and also

235
00:22:15,840 --> 00:22:21,800
add some certain custom dimension tracking with the given values that you have for the

236
00:22:21,800 --> 00:22:24,880
given departments and store it there.

237
00:22:24,880 --> 00:22:31,160
And then you are able to filter it, segment it, or create custom reports with it, as you

238
00:22:31,160 --> 00:22:32,160
wish.

239
00:22:32,160 --> 00:22:38,920
Yeah, it should be easily doable with the tag manager.

240
00:22:38,920 --> 00:22:44,960
Typically, you could just say, OK, there's this element on the page which is on the future

241
00:22:44,960 --> 00:22:48,640
on the footer who has those given HTML attribute.

242
00:22:48,640 --> 00:22:54,720
So I'm taking out this given attribute that I insert within the custom dimension field

243
00:22:54,720 --> 00:23:01,440
in the Matomo custom variable, configuration variable within the tag manager.

244
00:23:01,440 --> 00:23:06,120
And every time that the page is going to be loaded, as this given element is on the page,

245
00:23:06,120 --> 00:23:09,480
it's going to grab it and put it as a custom dimension.

246
00:23:09,480 --> 00:23:16,560
And then it will slice your categories accordingly.

247
00:23:16,560 --> 00:23:24,200
Perhaps also as an important comment on exactly this problem is, from the performance view

248
00:23:24,200 --> 00:23:32,000
of things, you should always try to put as much business complexity in the tracking runtime.

249
00:23:32,000 --> 00:23:39,560
So try to put as much complexity as possible from your data reporting and dashboard reporting

250
00:23:39,560 --> 00:23:42,340
and your segment definitions, et cetera.

251
00:23:42,340 --> 00:23:48,480
Try to put it into the tag manager, because if you have to filter for a custom dimension,

252
00:23:48,480 --> 00:23:59,200
it is very quick, it is very performant, very effective, and also the underlying MySQL queries.

253
00:23:59,200 --> 00:24:09,720
If you, for example, have your segment definition grouping 100 page URLs to try to define your

254
00:24:09,720 --> 00:24:16,040
branch or something else, some page grouping via segment, it will be very ineffective and

255
00:24:16,040 --> 00:24:17,040
slow.

256
00:24:17,040 --> 00:24:24,080
So try always to do as much as possible in the tag manager with custom dimensions, track

257
00:24:24,080 --> 00:24:34,120
some events with extra data, and so your performance stays good.

258
00:24:34,120 --> 00:24:46,960
I just noticed that I missed a question a bit above for Melanie.

259
00:24:46,960 --> 00:24:52,240
They have problems aggregating certain reports where monthly reports have fewer data than

260
00:24:52,240 --> 00:24:56,520
the daily reports in the internal search report.

261
00:24:56,520 --> 00:25:00,600
And which actions could we take to analyze the problem?

262
00:25:00,600 --> 00:25:06,760
So I think this is something that's hard to do a general guess, but how can one approach

263
00:25:06,760 --> 00:25:09,360
such an issue?

264
00:25:09,360 --> 00:25:14,520
What is the easy way to get in the direction of the issue?

265
00:25:14,520 --> 00:25:19,240
How to start troubleshooting?

266
00:25:19,240 --> 00:25:31,320
So perhaps I will just start on trying to answer that.

267
00:25:31,320 --> 00:25:40,640
The troubleshooting probably starts with making clear which part of the system makes this

268
00:25:40,640 --> 00:25:41,640
error.

269
00:25:41,640 --> 00:25:54,600
So probably one thing to keep in mind is that the archiving could fail on the monthly archives.

270
00:25:54,600 --> 00:25:57,280
Mostly this is the case when the memory limit is reached.

271
00:25:57,280 --> 00:26:05,200
So you will have to search in the server log files for errors in the archiving process.

272
00:26:05,200 --> 00:26:10,800
When you have a look, so if you're using the cron drop, so you should make sure that you

273
00:26:10,800 --> 00:26:13,600
disable the browser archiving.

274
00:26:13,600 --> 00:26:20,240
I think we can attach some links to that page later on in the chat also that you can make

275
00:26:20,240 --> 00:26:25,320
sure that you enable the cron drop for the asynchronous archiving of data.

276
00:26:25,320 --> 00:26:34,520
And then you can also read on that page how to log the things that this cron drop writes

277
00:26:34,520 --> 00:26:41,720
to the console, how to log it into a file and then analyze that file for errors in the

278
00:26:41,720 --> 00:26:45,680
monthly archive generation.

279
00:26:45,680 --> 00:26:53,080
That's probably in 90% of the cases the problem that the archiving job is failing for some

280
00:26:53,080 --> 00:26:54,080
reason.

281
00:26:54,080 --> 00:26:57,200
That's in general I think a good advice.

282
00:26:57,200 --> 00:27:05,440
If there are weird issues with the reports, maybe check the output of the archiving process

283
00:27:05,440 --> 00:27:09,360
if you're using a cron drop to regularly call it.

284
00:27:09,360 --> 00:27:14,360
Because it can be quite verbose and there are some parameters which you can add to make

285
00:27:14,360 --> 00:27:15,880
it even more verbose.

286
00:27:15,880 --> 00:27:20,800
And sometimes you can find something weird or some warnings in there which are a good

287
00:27:20,800 --> 00:27:25,000
hint in which direction to look.

288
00:27:25,000 --> 00:27:33,040
A short question from Niklas who has problems with the heatmaps before I will just quickly

289
00:27:33,040 --> 00:27:38,840
recommend you to send a quick email about it to the premium plugin support team.

290
00:27:38,840 --> 00:27:43,560
They can help you with this probably quite easily.

291
00:27:43,560 --> 00:27:49,360
Then another question from Chris because we talked about get and post requests previously.

292
00:27:49,360 --> 00:27:53,240
If there is a way to switch all requests to get.

293
00:27:53,240 --> 00:28:01,000
I can say something on that and by default the Matomo tracking uses send beacon from

294
00:28:01,000 --> 00:28:03,400
the browser which uses a post request.

295
00:28:03,400 --> 00:28:09,520
You can either disable this with the method disable always use beacon or something like

296
00:28:09,520 --> 00:28:15,880
that or you can force the request method with set request method.

297
00:28:15,880 --> 00:28:24,400
Those should be JavaScript tracking options you can send with a tracking code.

298
00:28:24,400 --> 00:28:29,680
The only thing to keep in mind there is of course there is a limit of how much data can

299
00:28:29,680 --> 00:28:32,720
be sent in the get request.

300
00:28:32,720 --> 00:28:39,120
Matomo also the old versions automatically switch to a post request if the older data

301
00:28:39,120 --> 00:28:41,720
that would be sent would be too large.

302
00:28:41,720 --> 00:28:47,400
I think it's quite a lot of data that would need to be sent for it to be too large for

303
00:28:47,400 --> 00:28:48,400
a get.

304
00:28:48,400 --> 00:28:53,680
But if the website you're tracking has enormously long URLs or something like this, this is

305
00:28:53,680 --> 00:28:56,400
something to maybe keep in mind.

306
00:28:56,400 --> 00:29:02,520
Just to mention it, this will always be the case if you are tracking heatmaps and stuff

307
00:29:02,520 --> 00:29:10,120
like that because it tries to send all the HTML body with a tracking request and that

308
00:29:10,120 --> 00:29:17,000
is required to send with post as get won't work.

309
00:29:17,000 --> 00:29:22,920
Another question by Valentin, which was a bit answered in the chat already, I just want

310
00:29:22,920 --> 00:29:25,680
to mention it here on stream for the recording.

311
00:29:25,680 --> 00:29:31,080
If it is possible to set the custom time for a scheduled task, for example, to choose that

312
00:29:31,080 --> 00:29:35,840
the scheduled task should always run in the night when there are fewer Matomo website

313
00:29:35,840 --> 00:29:40,600
visitors that the server has less load or something like this.

314
00:29:40,600 --> 00:29:44,040
And as far as I know, there isn't yet a feature like this.

315
00:29:44,040 --> 00:29:48,640
It might be something that could be added and that a lot of people might be interested

316
00:29:48,640 --> 00:29:49,640
in.

317
00:29:49,640 --> 00:29:55,480
A workaround that Tasman is suggested is, of course, you can run the Matomo scheduled

318
00:29:55,480 --> 00:29:57,440
task individually.

319
00:29:57,440 --> 00:30:03,480
So if this is something you care about, you could set up a separate Chrome jobs for all

320
00:30:03,480 --> 00:30:08,160
the scheduled tasks in Matomo.

321
00:30:08,160 --> 00:30:17,680
One question by T Scott is, are there plans to introduce filtering much like what exists

322
00:30:17,680 --> 00:30:24,480
in Google Analytics because segments don't cover all visits, which doesn't suit their

323
00:30:24,480 --> 00:30:25,480
purpose?

324
00:30:25,480 --> 00:30:39,240
As far as I know, there is actually an issue on GitHub, but I don't know the current state

325
00:30:39,240 --> 00:30:40,240
of it.

326
00:30:40,240 --> 00:30:43,360
But I guess it isn't planned currently.

327
00:30:43,360 --> 00:30:48,160
Yeah, I'm sharing the same answer.

328
00:30:48,160 --> 00:30:50,000
I don't have the answer here.

329
00:30:50,000 --> 00:30:55,880
So for the guys who are not familiar with Google Analytics, typically in Matomo, you

330
00:30:55,880 --> 00:30:58,280
get a different way of working.

331
00:30:58,280 --> 00:31:04,360
So they have like within the global settings, you can exclude by IP address and by adding,

332
00:31:04,360 --> 00:31:09,640
for example, the spam tracking prevention, you can exclude by countries, but you don't

333
00:31:09,640 --> 00:31:15,240
have as much flexibility as the Google Analytics filter, which are including those two dimensions

334
00:31:15,240 --> 00:31:21,600
that I just mentioned, including as well user agent, I forgot in Google Analytics, you have

335
00:31:21,600 --> 00:31:29,040
all the other dimensions, possibility to use as filter to come in, to filter out or include

336
00:31:29,040 --> 00:31:33,180
out within Google Analytics.

337
00:31:33,180 --> 00:31:36,960
And it's true that currently Matomo does not offer them all.

338
00:31:36,960 --> 00:31:39,240
And I don't know if they have plan of doing it.

339
00:31:39,240 --> 00:31:44,200
Yeah, I posted a link to the GitHub issue Stefan referred to.

340
00:31:44,200 --> 00:31:51,180
If you're interested in this topic, I'd recommend you to look it up there.

341
00:31:51,180 --> 00:31:56,240
In the meantime, while I'm looking for the chapter, one question I think would always

342
00:31:56,240 --> 00:32:04,400
be interesting to hear is, how can one contribute to Matomo?

343
00:32:04,400 --> 00:32:12,800
Either how can I report, notice a bug in Matomo, how can I report it, or if I know how to program,

344
00:32:12,800 --> 00:32:15,760
how can I even fix it myself?

345
00:32:15,760 --> 00:32:22,800
Okay, so it follows up, I think, a bit to the thing I mentioned in the beginning.

346
00:32:22,800 --> 00:32:29,400
So if you just have a vague issue and are not sure what is exactly going wrong, the

347
00:32:29,400 --> 00:32:32,240
easiest thing is to ask on the forum.

348
00:32:32,240 --> 00:32:38,200
And that way, find out a bit closer what exactly is the issue, what is going wrong.

349
00:32:38,200 --> 00:32:51,080
If you know exactly what is going wrong and how someone can reproduce it, then you can

350
00:32:51,080 --> 00:32:54,920
go directly to GitHub and create a GitHub issue for it.

351
00:32:54,920 --> 00:33:01,200
Of course, adding steps how to reproduce it makes it a lot easier to understand the issue,

352
00:33:01,200 --> 00:33:08,080
makes it a lot easier to fix it because oftentimes it's obvious for the person writing a GitHub

353
00:33:08,080 --> 00:33:16,040
issue to know, it's obvious for them what they are talking about, but if you're not

354
00:33:16,040 --> 00:33:19,960
up to date on what they are doing, it's quite hard to understand it.

355
00:33:19,960 --> 00:33:25,640
And adding steps to exactly reproduce the issue makes it a lot easier to understand

356
00:33:25,640 --> 00:33:35,040
the issue, and that way also a lot easier for a developer to debug it and then fix it.

357
00:33:35,040 --> 00:33:41,640
And of course, if you're a developer yourself and know how to fix it, then the easiest thing

358
00:33:41,640 --> 00:33:48,880
is of course to change the code, create a pull request with the change and contribute

359
00:33:48,880 --> 00:33:52,960
it to Matomo yourself.

360
00:33:52,960 --> 00:33:59,720
If you're a developer but don't know, not that familiar with Matomo, then you can also

361
00:33:59,720 --> 00:34:08,960
go to the Matomo developer page, which has a great page about how to get started with

362
00:34:08,960 --> 00:34:15,960
the code base of Matomo and how to set up a local Matomo development instance, which

363
00:34:15,960 --> 00:34:27,960
makes it a lot easier to contribute code changes.

364
00:34:27,960 --> 00:34:36,280
More analytics-related question about funnels, maybe for Stefano or Ronan, do you see the

365
00:34:36,280 --> 00:34:37,280
question yourself?

366
00:34:37,280 --> 00:34:40,880
I'll repeat it for people watching the recording.

367
00:34:40,880 --> 00:34:44,680
Talking about funnel refills, how is it working?

368
00:34:44,680 --> 00:34:49,760
If you have a funnel like this, where the first step is entry, then going to a conversion

369
00:34:49,760 --> 00:35:00,560
directly, does it get refills in the steps afterwards before the goal conversion?

370
00:35:00,560 --> 00:35:02,880
Maybe Ronan or Stefano have something to say about it.

371
00:35:02,880 --> 00:35:04,800
I'm not that familiar with it.

372
00:35:04,800 --> 00:35:06,800
I can't really help here.

373
00:35:06,800 --> 00:35:11,880
Otherwise, the question would again be to ask the support team.

374
00:35:11,880 --> 00:35:18,680
Well, you can of course set your goal in there and the conversion will be set in the second

375
00:35:18,680 --> 00:35:23,600
step when it's directly being achieved.

376
00:35:23,600 --> 00:35:31,160
It depends whether the fifth step is also being set as a conversion in your goal configuration.

377
00:35:31,160 --> 00:35:37,920
You can set it, set Matomo to make a conversion multiple times or to do it only once within

378
00:35:37,920 --> 00:35:38,920
a given visit.

379
00:35:38,920 --> 00:35:47,560
So that kind of things matter in your goal configuration, but as soon as someone moves

380
00:35:47,560 --> 00:35:54,680
across additional steps, it depends on whether certain steps are required or not to check

381
00:35:54,680 --> 00:35:59,080
whether the steps have been achieved and it really depends on your goal configuration

382
00:35:59,080 --> 00:36:03,200
settings there.

383
00:36:03,200 --> 00:36:15,720
A question by Marcel that I think a lot of people are asking is they have a website and

384
00:36:15,720 --> 00:36:22,240
they want to collect the reader feedback, so feedback by the people visiting the website,

385
00:36:22,240 --> 00:36:26,560
something like how did the story make you feel?

386
00:36:26,560 --> 00:36:32,240
It makes sense to collect this data in Matomo to connect it to the other data and could

387
00:36:32,240 --> 00:36:38,360
this be realistically implemented on top of Matomo or with Matomo as a data store?

388
00:36:38,360 --> 00:36:44,880
Of course, you could program something like this rather easily yourself if you do the

389
00:36:44,880 --> 00:36:51,200
user interface part yourself and then send the data, for example, as a custom event or

390
00:36:51,200 --> 00:36:58,840
something like this, but by chance, we have a talk tomorrow where Thomas Pearson is talking

391
00:36:58,840 --> 00:37:05,960
about a plugin they developed which does quite exactly this, which adds a feedback feature

392
00:37:05,960 --> 00:37:09,200
to a website using Matomo.

393
00:37:09,200 --> 00:37:13,000
So you might want to listen to this talk tomorrow.

394
00:37:13,000 --> 00:37:17,800
Yeah, I added the link within the chat.

395
00:37:17,800 --> 00:37:24,880
I think don't focus yourself on the form which is on the page link that I show you because

396
00:37:24,880 --> 00:37:31,160
I think that's in the link that you sent, you're like eight different buttons.

397
00:37:31,160 --> 00:37:38,040
I think that you can add with this plugin custom layout in order to have more than two

398
00:37:38,040 --> 00:37:40,440
buttons and to have eight like yours.

399
00:37:40,440 --> 00:37:45,320
The best thing is, yes, of course, to ask Thomas, but they work since probably two years

400
00:37:45,320 --> 00:37:46,320
on this plugin.

401
00:37:46,320 --> 00:37:48,520
So that's going to be a premium feature.

402
00:37:48,520 --> 00:37:53,000
So that's more the part where your hype go down, let's say.

403
00:37:53,000 --> 00:37:59,680
But it's always good to discuss with him to see, in fact, how they did it and the spirit

404
00:37:59,680 --> 00:38:00,680
behind.

405
00:38:00,680 --> 00:38:07,280
And I had the chance to assist to several conferences of demo that he made and it's

406
00:38:07,280 --> 00:38:18,560
very impressive and great feature.

407
00:38:18,560 --> 00:38:25,680
Maybe in the meantime, also a more marketing related or user related question that I get

408
00:38:25,680 --> 00:38:33,680
a lot of times is now that there is Matomo tech manager, what are the differences between

409
00:38:33,680 --> 00:38:41,760
using Matomo the regular way and using Matomo tech manager?

410
00:38:41,760 --> 00:38:46,520
So I could say a few words about that.

411
00:38:46,520 --> 00:38:53,400
So basically or generally both things are working and will work in the future.

412
00:38:53,400 --> 00:38:59,640
So that's the important thing to know at first.

413
00:38:59,640 --> 00:39:07,200
But mostly also with bigger teams working on Matomo, also more than one analyst.

414
00:39:07,200 --> 00:39:14,980
And if you have more than one requirement collected for tracking capabilities and everything,

415
00:39:14,980 --> 00:39:20,040
the tech manager makes more sense than the legacy tracking code.

416
00:39:20,040 --> 00:39:28,440
The reason for that is probably if you want to extend your tracking concept and you want

417
00:39:28,440 --> 00:39:34,400
to have more tracking features, collect custom dimensions, events, and have some behavioral

418
00:39:34,400 --> 00:39:37,760
stuff included into the tracking code.

419
00:39:37,760 --> 00:39:44,320
For example, if it's just a simple thing like tracking clicks on a call to action button

420
00:39:44,320 --> 00:39:51,720
or some kind of other interaction stuff, you would always have to have a release cycle

421
00:39:51,720 --> 00:39:58,640
and close contact to the developer of the website to just give him the code and prepare

422
00:39:58,640 --> 00:40:00,160
everything.

423
00:40:00,160 --> 00:40:02,120
And he has to include it in the website.

424
00:40:02,120 --> 00:40:06,040
And then you have to release a new version of the website to include the new version

425
00:40:06,040 --> 00:40:07,580
of the tracking code.

426
00:40:07,580 --> 00:40:12,800
And then in worst case, you have made an error anywhere or you need more data.

427
00:40:12,800 --> 00:40:19,400
You have a stakeholder that called with a feature request in the last second and you

428
00:40:19,400 --> 00:40:25,400
didn't bring it into the current sprint or something like that.

429
00:40:25,400 --> 00:40:30,080
So it's quite a nightmare to maintain this kind of code.

430
00:40:30,080 --> 00:40:37,520
With a tech manager code, it's mostly or most of the things you can do within the tech manager

431
00:40:37,520 --> 00:40:40,880
configuration in the generic user interface.

432
00:40:40,880 --> 00:40:48,960
So this is not a free ticket for everyone to do everything with tags and release new

433
00:40:48,960 --> 00:40:52,960
container versions and stuff because you can't break the site with that.

434
00:40:52,960 --> 00:40:55,320
So there should be a workflow for that.

435
00:40:55,320 --> 00:41:00,920
But basically, it is your time to market for the tracking is much better than with the

436
00:41:00,920 --> 00:41:03,000
legacy code.

437
00:41:03,000 --> 00:41:06,280
And also the maintainability is much better.

438
00:41:06,280 --> 00:41:12,320
And you can feed information that you can't collect from the HTML directly.

439
00:41:12,320 --> 00:41:14,660
You can gather via the data layer.

440
00:41:14,660 --> 00:41:21,560
So that's the single point of connection between you and the developer or the CMS is the data

441
00:41:21,560 --> 00:41:23,120
layer.

442
00:41:23,120 --> 00:41:28,120
You have to ask the developer to pass over information that you can get from the HTML

443
00:41:28,120 --> 00:41:33,200
and then you can pass it over to the tech manager and process everything with it.

444
00:41:33,200 --> 00:41:36,840
So it makes things in life much easier here.

445
00:41:36,840 --> 00:41:42,200
I also just noticed if you're really interested in this topic, you might want to leave the

446
00:41:42,200 --> 00:41:48,320
session right now and go to the other livestream room because Thomas Persson is currently talking

447
00:41:48,320 --> 00:41:53,920
about TomoTag Manager and the basics of it.

448
00:41:53,920 --> 00:41:58,960
In addition, you have also everything in one place.

449
00:41:58,960 --> 00:42:05,560
Every tracking script, you know where the tracking script is implemented and so on.

450
00:42:05,560 --> 00:42:13,920
And if you do a relaunch, you don't have to make a list of all the tracking scripts in

451
00:42:13,920 --> 00:42:16,640
page path 500 and so on.

452
00:42:16,640 --> 00:42:24,680
And so you can only open the tech manager container and see which scripts you are using.

453
00:42:24,680 --> 00:42:32,200
Sometimes it's easier to troubleshoot everything with the development or debug view.

454
00:42:32,200 --> 00:42:40,200
And you have some templates, some you can fill in so you don't have to code something

455
00:42:40,200 --> 00:42:43,240
if you're not familiar with coding.

456
00:42:43,240 --> 00:42:51,320
And so you can use some templates to integrate some external services.

457
00:42:51,320 --> 00:42:58,080
And if you want to roll back, you have a version control and I think it's easier to roll back

458
00:42:58,080 --> 00:43:03,360
a tech manager version than a release version.

459
00:43:03,360 --> 00:43:12,240
So I think in time, you can easily roll back a version in tech manager and like to say,

460
00:43:12,240 --> 00:43:20,240
you have to go to the developer and say, okay, can you reverse please the last version?

461
00:43:20,240 --> 00:43:22,320
And so it's easier.

462
00:43:22,320 --> 00:43:29,880
And yeah, sometimes you can see you have also documented who changes what.

463
00:43:29,880 --> 00:43:33,360
But okay, in the development, you have this too.

464
00:43:33,360 --> 00:43:38,960
But there you have the user permission management in one place.

465
00:43:38,960 --> 00:43:45,160
And I think some of this are all very good features.

466
00:43:45,160 --> 00:43:54,120
And I think in mind, I would never integrate tracking scripts without a container without

467
00:43:54,120 --> 00:44:02,680
Matomo tech manager or something else, something some other tech managers.

468
00:44:02,680 --> 00:44:14,440
Hey, I think I have answered most of the questions.

469
00:44:14,440 --> 00:44:17,360
Of course, the Matomo camp doesn't end now.

470
00:44:17,360 --> 00:44:21,960
We still have a rest of the day and a full day tomorrow.

471
00:44:21,960 --> 00:44:28,600
And if you have any more questions, just feel free to ask them in the chat.

472
00:44:28,600 --> 00:44:29,960
People will still be here.

473
00:44:29,960 --> 00:44:32,160
People will still be able to answer.

474
00:44:32,160 --> 00:44:39,280
And maybe if you want to talk to other people, again, the advertisement in the breaks or

475
00:44:39,280 --> 00:44:47,960
maybe after the last talk today, you can go to meet.matomocamp.org and talk to other attendees

476
00:44:47,960 --> 00:44:49,640
of Matomo camp.

477
00:44:49,640 --> 00:44:51,360
Maybe some of us will be there.

478
00:44:51,360 --> 00:44:53,360
I'm not sure.

479
00:44:53,360 --> 00:44:57,200
Well, thank you very much for all the questions that you ask.

480
00:44:57,200 --> 00:45:03,120
And we hope that you are having a great Matomo camp.

481
00:45:03,120 --> 00:45:04,120
Thank you very much.

482
00:45:04,120 --> 00:45:07,120
Have a nice evening, guys.

483
00:45:07,120 --> 00:45:09,520
You too.

484
00:45:09,520 --> 00:45:37,520
Bye bye.