I have attempted to add iOS and android tags on this for visibility for those teams but I might have the tag wrong. @JMinor and @Charlotte let me know if I got that right. Thanks!

Thanks @spatton for this ticket, and for laying out the asks so succinctly. It has been a great help. Also thanks @DStrine for the tags - yes, they're the right ones and they did help a lot!

@JMinor @Dbrant @JoeWalsh and I have consulted, and herewith our thoughts:

App article fundraising: This could be big. Josh Minor says it's negotiable for iOS, and already on the roadmap for Android (ty Charlotte). Here's our current best 'small' mobile banner for reference.

This is a possibility. @Dbrant has already done some initial exploration with @Seddon and ruled out using CentralNotice, but @JMinor has a couple of other ideas worth exploring, and he will file a ticket for this.

Weekday variable: Using copy that includes weekday has proven to be very effective. This would require a simple variable (%weekday%) that outputs the current day of the week. We’re using it throughout our banner web portfolio.

Not this year, I'm afraid. We'd need to have multiple versions of the app to support this, as it wouldn't be backwards compatible, and the in-article ads take priority.

Device variable: In the mobile small banner, we’re using the following copy: „Donating is easy on your phone and only takes a minute.” Replacing the phone part with the reader’s device name (Galaxy S9, Galaxy Note, iPhone, iPad, etc.) will almost certainly be effective.

Ditto weekday variable.

Visual helpers [for Android]: In 2018, we got the red border from our current best web content into iOS; would it be possible to include in Android, too?

Do-able. I have filed a ticket which you will see as a subtask for this ticket.

Hey @JMinor, flagging that I've updated the original task to include an ask for impression data for iOS; we can already get it for Android. Just tell me if I should have logged this somewhere else, instead! TY.

Posting updated copy in this thread for the use of anyone working on the iOS or android implementations.

6 variants for the 6 countries in which these banners will run.

US
To all our readers in the U.S.,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $16.36, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

UK
To all our readers in the UK,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about £10, but 99% of our readers don't give. If everyone reading this gave £2, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

IE
To all our readers in Ireland,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about €10, but 99% of our readers don't give. If everyone reading this gave €2, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

AU
To all our readers in Australia,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $15, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

NZ
To all our readers in New Zealand,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $15, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

CA
To all our readers in Canada,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $15, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

@cooltey and @schoenbaechler - FYI in case you've not seen this:

In T235447#5665491, @spatton wrote:

Posting updated copy in this thread for the use of anyone working on the iOS or android implementations.

6 variants for the 6 countries in which these banners will run.

US
To all our readers in the U.S.,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $16.36, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

UK
To all our readers in the UK,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about £10, but 99% of our readers don't give. If everyone reading this gave £2, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

IE
To all our readers in Ireland,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about €10, but 99% of our readers don't give. If everyone reading this gave €2, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

AU
To all our readers in Australia,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $15, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

NZ
To all our readers in New Zealand,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $15, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

CA
To all our readers in Canada,
It's a little awkward, so we'll get straight to the point: Today we humbly ask you to defend Wikipedia's independence. We depend on donations averaging about $15, but 99% of our readers don't give. If everyone reading this gave $2.75, we could keep Wikipedia thriving for years to come. The price of a cup of coffee today is all we need.
Wikipedia is a place to learn, not a place for advertising. It unites all of us who love knowledge: contributors, readers and the donors who keep us thriving. We know that most people will ignore this message. But if Wikipedia is useful to you, please take a minute to help keep it growing. Thank you.

Hey @AndyRussG, here's the parent task for the December app fundraising banners. Is it possible to reuse CentralNotice beacons to count impressions of our app banners? If so, what do we need to do? TY!

Hi all! Congrats to all for your work on this...

In T235447#5673509, @spatton wrote:

Hey @AndyRussG, here's the parent task for the December app fundraising banners. Is it possible to reuse CentralNotice beacons to count impressions of our app banners? If so, what do we need to do? TY!

It's not impossible, since it's just a call to the servers with info sent as URL parameters, which is tracked existing via Analytics systems for all HTTP request, plus a few extra, specific bits and pieces we have in place.

However, I wouldn't recommend using it, for a few reasons:

• The existing beacon is essentially a hack. We're in the process phasing it out and replacing with a new data pipeline that will use proper infrastructure that Analytics uses for tracking events, namely EventLogging.
• Analytics' EventLogging infrastructure has several advantages in terms of data processing and scalability that the existing beacon lacks.
• Mixing app events into the existing beacon stream could make it more complex to query data about CentralNotice events, especially if we need to track banner displays in relation to Web page views.

Instead of using the existing beacon, it should be simple to create an EventLogging schema with the appropriate data points, then send a conforming request to an EventLogging endpoint.

I guess the first question I'd have about this is, how are events generally tracked in the apps? Is there any reason to use something different? Maybe the apps already send EventLogging events?

A second step for this that comes to mind could be to integrate app banner events into the new data pipeline that we're building. Again, this will replace the existing beacon, as well as the old, fragile, hard-to-maintain FR-specific bits server-side.

Maybe I'm missing what the motivation might be to use the existing beacon, though? Please let me know if there are requirements here that I've missed... Hope this is helpful!!! Thanks!

@spatton - As @AndyRussG mentions, the app already sends EventLogging events, and we should try where possible to re-use existing structures rather than hacking things together. If you can cut a ticket on the Product Analytics board for creating the necessary schema, detailing your requirements, @kzimmerman can prioritise accordingly.

Kate - this is for a campaign that begins 2 December - tracking impressions on a fundraising feed card and in-article advertisement. Sam might have other particular things he wants to measure that perhaps you or someone on your team can discuss with him? (We can presumably re-use much if not all of what we did for tracking last year - perhaps @Dbrant can weigh in once Sam's written down his requirements.)

Thanks for the detailed response @AndyRussG and for weighing in too, @Charlotte. I agree re: re-use of existing structures.

Our understanding is that we can already get app fundraising banner impression data via the EventLogging that was present in the December 2018 campaign - @JoeWalsh shared a file when we spoke last week.

If that can be confirmed, then our basic need is met 👍

In T235447#5678883, @spatton wrote:

Thanks for the detailed response @AndyRussG and for weighing in too, @Charlotte. I agree re: re-use of existing structures.

Our understanding is that we can already get app fundraising banner impression data via the EventLogging that was present in the December 2018 campaign - @JoeWalsh shared a file when we spoke last week.

If that can be confirmed, then our basic need is met 👍

Cool. @Dbrant or @cooltey - can you briefly check over the existing schema and make sure we're good to go with impression numbers for both the card and the in-article thingy?

@Charlotte
Here's the donate URL that contains logging parameters:
https://donate.wikimedia.org/?uselang=en&utm_medium=WikipediaAppFeed&utm_campaign=Android&utm_source=app_201812_6C_control

The only thing we need to change might be the value of utm_source: app_201812_6C_control -> app_201912_6C_control, or others you'd like to input.

Thanks @cooltey - @spatton do you want to change anything but the year? Not sure what your naming convention is.

Thanks for the ping @Charlotte; I've posted an updated URL in both the android and iOS specific tasks, but will add both links here for the sake of completion:

iOS: https://donate.wikimedia.org/?uselang=en&appeal=JimmyQuote&utm_medium=WikipediaApp&utm_campaign=iOS&utm_source=app_201912_6C_control

Android: https://donate.wikimedia.org/?uselang=en&appeal=JimmyQuote&utm_medium=WikipediaApp&utm_campaign=Android&utm_source=app_201912_6C_control

It looks like you're using existing tracking, @spatton; is that right? No need for someone on Product Analytics to weigh in?

@Charlotte @JMinor @kzimmerman @spatton for impressions, we could use the existing schema and treat them the same as last year (as explore feed card impressions). We'd lose the distinction of in-article impression vs feed impression but if you're OK with that, it'd be easier than adding a new schema at this point

[Edit:] it looks like on iOS's MobileWikiAppiOSFeed we could vary the "label" based on whether or not it was in the article or the feed and on Android's MobileWikiAppFeed we could add a separate "cardType". Let me know if you'd like to go this route or keep it uniform with last year's cards.

@JoeWalsh I was thinking we'd use the version to distinguish placement, but if there's a simple label field that sounds like the way to go.

Just to sum up, we'll use existing event logging schemas for tracking feed content, with the addition of the placement type in the "label" field for iOSFeed (iOS) and "cardType" for WikiAppFeed (Android).

For FR folks, you can pull directly from that schema and/or the apps folks can work with you to get the numbers.

Great, thanks Josh. I should be able to get that data myself, will let you know if I have any trouble.