Recommend method for recapturing addressing organic search
Categories
(Data Science :: Investigation, task, P1)
Tracking
(Not tracked)
People
(Reporter: harter, Assigned: bmiroglio)
References
Details
Ben did some great analysis in Q4 describing organic search and how it correlates with custom homepages and newtabs. Let's refine this work further so we can motivate features to recapture this volume.
A couple of ideas:
- It looks like Custom URL and Extension Homepages are creating an increase in organic search. It would be great if we could refine this
- What extensions are changing homepages? Do we think this is an intentional change or hijacking?
- What Custom URLs are being set? This will be more difficult to do while preserving privacy but maybe we can use some type of rough categorization. Or, maybe we can just ask users if they want to support Firefox by resetting their homepage URL if we detect they have a custom URLs that look like a search engine
- Can we identify any addons associated with an increase in organic search? Is that virtuous or hijacking?
I'm sure there are other avenues we could pursue. We don't need to investigate all of these now. The goal for Q1 is to motivate a feature to recapture some organic search. We can split off new bugs for any ancillary project work we'd like to do in the future. The actual feature launch can happen after Q1.
|
Reporter | |
Updated•5 years ago
|
|
||
Updated•5 years ago
|
|
Assignee | |
Comment 1•5 years ago
|
||
Per a conversation with harter, the ultimate report to come out of this bug will focus on the share of organic searches to tagged SAP searches.
Specifically, the plan is to quantify the share of organic searches to (tagged_sap + tagged_follow_on) searches. This is the most direct comparison to organic, since organic lumps in follow-on searches, while the standard sap field does not account for them. When talking about the "potential search and revenue increases" when recapturing organic search, this ratio is the clearest lens.
ni-ing harter to verify I have this correctly summarized.
|
|
||
Comment 3•5 years ago
|
||
Is this ticket still needed? Is this a request from within Data Science (Harter) or is there another stakeholder requesting this?
|
|
Assignee | |
Comment 4•5 years ago
•
|
||
From my perspective, we can close this bug, since this is now an ongoing issue, but there's not a clear path to recommending a single method. This is more of a whack-a-mole problem at this point, going on a case-by-case basis when something comes up in analysis or add-on reports.
Ryan, let me know if you disagree.
|
|
Reporter | |
Comment 5•5 years ago
|
||
I think this is resolved. We've made a few recommendations and actions this year so this should have been closed a while ago.
Description
•