Closed lukemulks closed 9 months ago
This issue is stale because it has been open for over a year with no activity. Remove stale label or add a comment to avoid this being closed in a weeks time.
This issue was closed because it has been inactive for a week since being marked as stale.
This issue is a follow-on to https://github.com/brave/ads-ui/issues/276
With some time passed since the release of Sponsored Images, I've observed a pattern both directly (through BD deals: Newegg, Verizon, Chess.com and through affiliate discussions) and indirectly (via discussions with the Sales team) that's having a negative direct impact on revenue.
This topic has been discussed for months, but the current solution being discussed will not adequately cover the needs to generate maximum revenue.
The issue at hand is reporting capability with Sponsored Images.
Sponsored Images are currently the highest value unit that the Sales team can sell in-market, and is a valuable lever for BD deals to gain traction with media partnerships and brands.
We currently do not have any accurate reporting capability for Sponsored Images, aside from estimates backed out from opt-in users.
Those estimates did not really work very well in the short term, and the near and long term impact on revenue is significant.
We have an ad unit that can display for everyone, and not just everyone, but for users that typically shy from commercial messaging. This is of high value. Brands have confirmed this, but are unable to justify the spend when we offer them zero reporting.
As a result, this premiere unit becomes significantly discounted.
I'm observing what should be a $1M/mo unit be discounted down to remnant and CPA rates because we have no event reporting instrumented. This is an alarming course, so I'm raising this issue after observing repeat discussions where the only apparent path forward is to either:
1) estimate based on opt-in user rates, which will not be a clear/accurate measurement of actual counts
or
2) rely on secondary reporting from brands, which can be even more problematic given their limited understanding of what is and isn't blocked by our platform.
I've run into issues with both eToro, Newegg and Lokai when attempting to go through the secondary reporting route.
Unfortunately, estimates from opt-in rates are not enough to satisfy brands and partners. To be competitive for advertising spend, accurate upper and mid-funnel reporting in the dashboard is required.
To be very clear, the ask is not for conversion reporting, or anything beyond country-level view and click counts on the unit.
Here's the specific ask:
Report view counts for Sponsored Images. This could be push counts from the server, or render counts from the client, but either way, should be a "dumb-count" signal back that does not include any PII or data other than country and
1
(has displayed) or0
(has not displayed).Report clicks from the Sponsored Image icon in the lower left corner of the unit. These could be "dumb clicks" that only report a
1
was clicked or0
was not clicked. No additional data other than country code. No URLs reported back. Only a simple engagement event.Reported events should display in the Brave Ads dashboard for advertisers. Both clicks, and views, and by country.
I would suggest we use P3A for this, but am unclear whether the P3A reporting would be statistical averages, or actual counts. Unfortunately, the prior will not suffice.
The other concern around P3A is that it is not currently live on mobile. Parity for desktop and mobile will be required in order to sell this unit at the value it can demand in-market.
I'm raising this issue now out of concern that I'm observing this unit being discounted so much that it's becoming comparable to "added value" bargain bin publisher placements that I used to run in my gig prior to Brave.
Based on experience with partners, prospects and internal meetings, the single greatest barrier to maximizing revenue and time/opportunity costs for this unit appears to be an unwillingness to instrument basic binary event counts, applicable to all instances where this unit could be displayed or engaged with.
I hope that this does not come off the wrong way, but every time that this item has been brought up in meetings, the response has been that we can only run report counts from opt-in estimates, which unfortunately will not suffice to be competitive and maximize value.
Capturing in an issue as an alternative approach, with the hope that stating a clear case here will help get a solution implemented that provides the best chance for the greatest value and success for this unit.
This unit is special. It's high-value art. It's been received well by people and prospects. We should be demanding that value. With the three items above, we can.
@jsecretan @BrendanEich