@gabriel-wmde could you please:

• broaden the buttons in ctrl so that the form behaves like in var (being a two-liner)
• check the free field in var. if I type in an amount on the landing there is no amount selected
• if possible make the space between radion button and button text a little little little higher?

@Tonina_Zhelyazkova_WMDE thanks, now everthing seems fine!

@Tonina_Zhelyazkova_WMDE: There is one data field missing I need to identify imps (viewport-sizes) on de.wikipedia.org. It's the field "webHost" (The current export only has the field "wiki" but if that field states "dewiki" this can also mean that the banner was on de.m.wikipedia,org.). The field "webHost" was included in the exports of the ticket linked in the task description. Could you please do another export?

looks good! thanks @Tonina_Zhelyazkova_WMDE

looks good. thanks @Tonina_Zhelyazkova_WMDE!

@Tim_WMDE I'm sorry, there is another mistake. Instead of the field 'viewportwidth' the field 'viewportheight' is exported twice. Could you please make another export?
Also, if it doesn't make much more effort could you include the viewport-tracking again ('bannername' = 'viewport_tracking')? (Originally this data was exported, too, but than cut to make the export smaller.)

looks good, thanks!

The data is now consistent with the older export! Thanks!

The export has an error. Instead of the field [day] the field [hour] was exported. Can you please provide another export? (hour can be kept, just add day)

@kai.nissen @Tim_WMDE This ticket is not done, yet or am I wrong? Could you please make another export?

@Jan_Dittrich I was wondering about these form names, too, but in Matomos forms view (plug-In Form Analytics). Than I realized that these are the forms of the old "10h16" donation page. It's a two pager; that's why Matomo created two forms in Form Analytics (which unfortunately makes it difficult to compare the old and new donation pages / forms). Perhaps in the goal view form 1 and 2 refer to the same thing.

@kai.nissen @Tim_WMDE
I'm sorry I need some more data fields in the export:

• opt_in
• receipt
• gender
• banner_impressions
• total_banner_impressions

@Tonina_Zhelyazkova_WMDE looks good! One last thing: The test is supposed to start today. Can you change the date in the tracking codes accordingly?

@kai.nissen @Tonina_Zhelyazkova_WMDE I think the equality of banner sizes is totally ok.

@Tonina_Zhelyazkova_WMDE looks good! Test can start tomorrow.

@gabriel-wmde two last things:

@gabriel-wmde Looks good but I noticed a few features (windows, firefox):

In T211858#4823614, @Tobias_Schumann_WMDE wrote:

I think I found another thing: The sentence "Wikipedia bringt uns alle, die Wissen lieben, zusammen: Beitragende, Lesende und die Spendenden, die uns finanzieren." is supposed to be added at >= 1900px according to description. It appears to me that it is already added at >=1400px. Can you please check?

@kai.nissen @Tim_WMDE
I think I found another thing: The sentence "Wikipedia bringt uns alle, die Wissen lieben, zusammen: Beitragende, Lesende und die Spendenden, die uns finanzieren." is supposed to be added at >= 1900px according to description. It appears to me that it is already added at >=1400px. Can you please check?

Sorry for being unconcrete @Tonina_Zhelyazkova_WMDE When I simulate Ipad and IpadPro with chrome dev tools the button looks like this:

@Tim_WMDE Here are some things I've noticed:

@Tim_WMDE just being extra cautious: with the second bullet point ("bisschen" --> "wenig") @tmletzko is referring to both banners.

@kai.nissen @tmletzko Buttons were supposed to become higher but not wider (to an extend in which that makes sense layoutwise). I would say it depends on time and effort needed to adjust that if that still shall be done.

@kai.nissen I've just changed the description and added the new sums for the banner variant. It will feature 8 buttons instead of 6.

@Tim_WMDE I've just finished the analysis. Here are the sums for the donation buttons:

@kai.nissen Unfortunately I only notice now: The description of that task is wrong. As the task title correctly states the export is about mobile test 2. However the banner names provided in the task description belong to test 3. So could you please make another export of data of test 2 for the dates: 12.10. - 11.11. 2018

@kai.nissen Sorry for posting this here, but I noticed the description is wrong: The campaign name in section general should be "C18_WMDE_mob04" and not "C18_WMDE_mob03"

@Tim_WMDE This should be added to all 4 banners.

In T207558#4742432, @tmletzko wrote:

also, please add the parameter "link" to the url of the MIttelverwendung as we do on mobile banners in order to not get in conflict with the donation tracking.

@kai.nissen looks good! Only one thing: in VAR "Heute beginnt unsere Spendenkampagne." should be "{{campaignDaySentence}}", I guess.

@kai.nissen the text of both banners needs to be adjusted:

@kai.nissen Please make the two sentences in page 1 in different lines, e.g. make a text break (?) after "Heute beginnt unsere Spendenkampagne."

@Tim_WMDE Thanks! Before we forget that on Monday: You can prolongue test 10 to Tuesday, Nov. 13th.

@Tim_WMDE Thanks!
The test will start today. But can you please change the start time to 4 pm (=3 pm utc i guess)? Important: You'll have to prolongue the ongoing test 9 to 4pm, too.
And furthermore please set the imp limit on test 10 to 5%.

I'm sorry @Tim_WMDE, I missed a thing: could you please make the salutation "Liebe Leserinnen und Leser, " not-bold in both banners? Thanks!

Hi @Tim_WMDE Now everything looks good.

Hi Tim, that was fast! I found some issues which need to be changed:

@Tonina_Zhelyazkova_WMDE Regarding the banner-close events of the mobile test I compared the data to last year and must correct me: the data is not unplausibly high.

@Tonina_Zhelyazkova_WMDE Thanks for the export. I have questions regarding the export of the "banner-closed" events:

• There is no data for destop tests 1-6 in the export and the earliest timestamp is Ocotber 12th. I guess you used incorrect dates in your query. The start date of the export is July 12th (see descpription)
• The banner-close events for mobile test 2 are very high and the percentage closed banners of banner impressions is so high that it is unplausible. Can it be that 10% instead of 1% of the events are being tracked? (but the export column "event_eventRate" says "0,01")

@Tim_WMDE I need that data already during this sprint, sorry - except for the data of test 8, of course. Perhaps we split the test 8-part to the next sprint? Data for the other tests could be gathered earliest the day after the start of test 8 (so that test 7 data is complete) which would be Thursday if all holds well.

@kai.nissen Since the test on mobile started today it would be good to check on Monday if data is being tracked properly. No need for an export, just a quick look if (plausible) data is been stored. Just a matter of caution based on experience of live ;)

@Tonina_Zhelyazkova_WMDE Yeah, i know. But these CN settings are not all bullet proof as I remember, but more a mixture of device / Wikipedia version (but i always forget if that affects mobile-->desktop or the other way around). Therefore ping @kai.nissen

@Tim_WMDE @kai.nissen tracks are ok, thanks.
i noticed another thing: when the var banner is viewed on desktop version of Wikipedia is disappears after opening it, e.g there is no "page 2". (ctrl banner works properly on desktop version) i know this banner is not meant to be shown on this version but if there is a possibility that a user somehow manages to see that banner on desktop version the problem should be fixed, i guess.

@kai.nissen ok!

@kai.nissen @Tim_WMDE can you please change the dates of the tracking codes to tomorrow's start date?
And another remark (sorry that late): the ctrl banner is a bit higher then the variant. is it still possible to adjust heights so that both banners have the same height? if not it's not such a big problem and the test can start tomorrow wothout a change.

@Tonina_Zhelyazkova_WMDE ok, thanks!

@Tonina_Zhelyazkova_WMDE the test can start today as planned (yesterday I wrote @gabriel-wmde that I was hesitating because of the number of donations in the current test), but can you please change the tracking codes to todays date? Thanks!

@Tonina_Zhelyazkova_WMDE Looks good, thanks!

@Tonina_Zhelyazkova_WMDE @kai.nissen
I found a bug in both banners. On firefox (windows 10, firefox quantum 60.2.1esr) on smaller view ports the donation form overlays the banner text. On chrome everything is fine.

@kai.nissen This task might be deleted, i guess. It's obsolete by the development of the responsive banner

Ok, we found the reason. I was erroneously looking for unique visitors on stats.wmde.de where I should have looked for Seitenansichten.

@kai.nissen I noticed that, too. After a few tries it provided data. Perhaps requests are not logged because of traffic load? (<-- pretending to understand technical background)

@kai.nissen @gabriel-wmde @Tim_WMDE @Tonina_Zhelyazkova_WMDE
I've just compared the now fixed data of tracking.wikimedia.de and stats.wikimedia.de Unfortunately the data still differs substantially. Of course a certain difference is expectable because in case of tracking.wmde.de only 1% is tracked. But the difference i see now is much higher than it used to be in 2016 when there was a parallel tracking on the two matomo (piwik in these days) instances as well. please see attached pictures:

@Tim_WMDE That's the way to go :)

@Tim_WMDE everything looks fine! test shall start on friday, august 17th.

@Tim_WMDE everything looks fine! test shall indeed start on friday, august 17th.

• Please display no donation progress bar (like in reference ctrl banner of 2017)

For the record: We decided to go with the solution to have a separated eMail form with opt-in text below. See:

@Tim_WMDE The unique page views did indeed increase substantially. Let's wait a few more days to see how the data is developing. Maybe it makes sense to have a deeper look into Matomo settings next week.

@Tonina_Zhelyazkova_WMDE looks good, thanks!

@Tonina_Zhelyazkova_WMDE @gabriel-wmde Hi, i just checked data in Matomo and it seems like the data is not tracked correctly. Daily Unique Page Views ("Eindeutige Besucher") are with values of ~300 much lower than the unique page views measured before by the Matomo of spenden.wikimedia.de (~800-1200). (I guess in both cases only 1% of the page views is being tracked, correct?) Can you please have a look soon?

@Tonina_Zhelyazkova_WMDE yes, would be cool if you can enable the new campaign and end the old campaign at 2 pm because than i have time to monitor incoming donations. at 3 pm i have to leave.

@Tonina_Zhelyazkova_WMDE could we start the test a little bit earlier tomrrow? than i am able to have a look at it, too. would 1 pm work for you? or - if that's your midday break - 1:30 pm or 2 pm is also fine. thanks!

@Tonina_Zhelyazkova_WMDE Now it's fine! Same size on different view ports.
I wrote Kai an E-Mail today so he should be aware.

@Tonina_Zhelyazkova_WMDE thanks! looks good, but one remark from my side: both banners are supposed to have an identical height (as far as possible with the different text). but var (with the shorter text) is instead higher than ctrl (with the longer text). can you please have a look at that?

@gabriel-wmde "in part" meens not for every banner (not for var today). but for ctrl the data of today is 23 unique visits according to matomo and 27 donations according to backend. just check the data. yesterday there are 39 unique visitos each for ctrl and var but 45 respectively 46
i don't realized any pattern, though.

@kai.nissen @gabriel-wmde @Tim_WMDE There is no data of landing page visits in the analysis tool of the backend. There are visits being tracked in matomo but they seem buggy as well: there are in part less visits than donations.

@Aklapper. ah yeah, thanks!

@Tim_WMDE @gabriel-wmde The banners look good! When tracking via EventLogging is implemented we can start the test. However, before tracking codes need to be updated.

After thinking about pros and cons of an impression tracking only when the banner is in fact displayed we decided - for now - to keep the tracking as it is. More than the exact information of real banner impressions we are interested in the correct sample size of the test group (and users that don't see banners are still part of the a/b test).

To add another idea: We could also add the E-Mail form field to the anonymous section (the part on the right side which becomes visible after choosing anonym). Ok, that would be contradictory to the notion of anonymous donations. But it would come with some advantages:

• There would not be the need for a separate E-Mail section in the donation form; The E-Mail form fields for private and corporate donations could stay where they are (more or less; they would need to move down to the end of the address section and the opt-in would be places benaeath)
• Without seperate E-Mail section we would not risk to decrease email-rates from donors with address (private and corporate donations)

@GoranSMilovanovic Ah, I thought the analysis was campaign-wise. Ok, looking forward to see the result.

Quoted Text My apologies to @Tobias_Schumann_WMDE for not pinging him earlier.

Don't worry :)

Hi @GoranSMilovanovic and @Jan_Dittrich. I've got some comments and questions on the analysis.

Hi @GoranSMilovanovic I'm sorry very much, that was just a (fatal) typing error. I wanted to say: "The reason to say the new landing page underperforms ..." So yeah, I saw decreased opt-in rates, decreased female salutation rates and increased rates of donations without address - with the new landing page.

Some more information about tracking keywords. It is unfortunately in one case not old/new landing page (lp) = ctrl/var lp. Please see this overview:

@Dzahn
Great, now it works! Thanks a lot!

@Dzahn Sorry to open that again. I do not succeed to log in into Pivot. Could it be that this has something to do with the unclarities around my nickname?

Hi @Peachey88 thanks for the hint! It wasn't, but now I created the link. I thought the linkage in LADP is enough. Thanks again.