Topic on Talk:New Editor Experiences

There are any number of ways to classify and think about the ideas we're evaluating—and, more importantly, to organize the recommendations that will come out of the New Editors process. But I'd like to argue for a way that really hit me as we worked through the user journeys exercise in our last session.

As I read the various user stories, similar themes emerged and coalesced in my mind. In particular, people had often faced similar challenges—and been helped in similar ways. With this in mind, the approach that will be most useful for our purposes, it now seems to me, is not just to evaluate which ideas are the strongest, or to parcel out different solutions that address different user personas, or to think about which assignments might fall into the wheelhouses of which WMF teams... Instead, I think we need to identify the key junctures in the New Editor journey, and then devise solutions aimed at addressing the users' needs at each phase.

Here are some examples of what I mean, just as straw men to illustrate what this might look like:

Juncture #1: Registration

• Ideas around onboarding, gathering profile information, welcoming, encouraging with recognition...

Juncture #2: exploring and finding a niche

• Ideas around progressive task suggestion, building awareness of the community, steering to Wikiprojects...

Juncture #3: first conflict/rejection

• Ideas around peer-to-peer assistance, improved policy pages...

The danger of not thinking in this way is that we run the risk of creating interventions that move users down the retention road a mile, only to drop them straight into the next (predictable) pitfall.

I'd always known that we wouldn't fix the problem of editor retention with just one intervention. But thinking about the new editor journey provides, I think, a useful framework for picking the most promising interventions and imagining how various efforts might fit together into a larger program of mutually-reinforcing parts. It's a way to turn a market-basket of ideas into a story about about what our users need and how we plan to help them. And, finally, asking which ideas help users most at particular points gives us more concrete criteria for evaluating and comparing very disparate ideas.

Re: Junctures: Yes. - I'm sure I've seen an infographic/project that demonstrated this idea quite well, but I cannot currently find it on mediawiki/metawiki.

The things that I found whilst looking, which are close but not exact matches, include (these are just the highlights, and should be semi-glanceable!):

• File:E2_Planning_-_Q3_2013.pdf&page=14 - Engagement Funnel
• m:Editor Growth and Contribution Program/Taghreedat#Workflow
• m:Editor Growth and Contribution Program/Contribution portal - Phase 0#Tutorial Pages
• m:Editor Growth and Contribution Program/Geo-targeted Editors Participation#Framework
• Growth/Prioritization
• Growth/2013-14 Goals#Our product framework
• File:Direct-to-page_onboarding_workflow_overview.pdf&page=2
• m:Research:Wikipedia Workflows/Article#Diagram
• m:File:Multimedia-Map-Workflows-August-08-2013.png
• Article Creation Workflow#/media/File:Article Creation Workflow diagram.png

Other related links I came across (whilst looking through search results for "funnel" and similar), as examples of some good prior work:

• Juncture #1: Registration
  • Editor campaigns/Technical design#Use cases and stories
  • Account Creation Improvement Project#Current account creation flow
  • outreach:Account Creation Improvement Project/Assumptions
  • Wikimedia Foundation Design/Login
  • Anonymous editor acquisition#Current user experience
  • m:Research:Account creation UX
  • Anonymous editor acquisition/Signup invites#Option 1: Pre-edit CTA

• Juncture #2: exploring and finding a niche
  • Onboarding new Wikipedians#User experience

• Juncture #h: Help gadget
  • New user help gadget

• Juncture #n: Post-first-edit
  • Post-edit feedback
  • m:Template A/B testing - fine-tuning the talkpage-warning template-messsages
  • m:Research:New editors' first session and retention#Summary
  • m:Research:Improve your edit

Lastly, this is still the best 10,000' overview (imo), albeit from 2011:

• Feature map

Hey Joe, FWIW, I agree with you 100%. It's different kinds of users at different junctures needing different things. Focusing on one at a time seems like an interesting approach, and i think focusing on specific user types and just a few junctions will serve us well. But we should be very careful, as you said, about what we choose to avoid opening up one door to help a user to find the next pitfall. The other thing that's been on my mind recently is how this isn't a funnel, but an ecosystem with many different components. If user type B moderates user type A, you can't 10x A without first 10Xing B's capacity (either through recruitment or better tools).

I agree on opening a door and then fall into an unexpected pitfall. From what I know, both from reading the interviews and personal experience, the big pitfall is the lack of human interaction: find someone to help, assist, discuss with, review...

Also, some recommandations we will publish can be handled directly by communities.

I agree these are relevant stages, and they seem useful to organise and evaluate ideas around them. I think that the stages identified are right, but I'd try to frame them in a way that is less coupled with technical solutions (such as registration) and have more clear connection with user goals (e.g., is the ultimate goal for a user to just "explore" or does such exploration has a purpose?).

I tried to adjust the language for the stages to reflect the above ideas:

• Feel invited to contribute. Covering registration process but also how we surface editors that an article is created by people like them and making them interested in joining.
• Find where to contribute in the best way. Surfacing interesting ways to contribute that are suited for their expertise, and new editors feel ready to accomplish (knowing not only how to use the tools to do it, but the key policy aspects that make a good contribution).
• Avoid issues and recover from them. Getting help and guidance from the system and/or humans when they experience a conflict or rejection of their work.

I agree also in being aware that the different stages are connected (you don't want to massively increase acquisition if you have big problems to retain those users). However, I also think we should not aim for the next stage to be perfect before doing any single improvement to the previous one, since that can also add unnecessary blockers. For example, if we have a simple idea to improve acquisition, I'd prefer trying it and deal with the potential case of becoming too successful (e.g., exposing it to a lower number of users) rather than postponing it in advance because the rest of the system is not ready for a potential huge influx of users (and not learning that the idea works so well for that stage).

It's a great idea to model the 'new editor journey'. Hopefully resources that help new contributors at each step can be 'plugged in' to this model. I suggest brainstorming (on-line) with some new editors.

Based on my newby experience, new contributors (including editors) don't go through the same learning process (journey). They certainly don't go through the steps in the learning process in the same order. Each new contributor follows his/her individual learning path (based on curiosity, immediate learning needs, the links he/she comes across, the editors he/she interacts with, etc.). The path that a translator or copy-editor takes will both be different to someone who joins to create a new article.

As an aside: the current WP learning materials/guidelines are IHMO not very suitable. They consist of a vast collection of long, text-based WP articles. For example, there are 367 "How to ..." articles". Many articles/guidelines are difficult to read, digest, remember and find again. Navigating between related pages (links, categories) can only be done by scrolling down to the bottom of each page.

One of the relevant opportunities in the REBOOT report was to "provide just-in-time guidance to new editors based on the tasks they want to accomplish". I think this would be a big improvement. There are very few Wikiproject training materials on YouTube and all pre-date the visual editor. I think there are opportunities to add training videos and on-screen tips (video/text).

In parallel with developing a model that includes all individual 'new editor/contributor journeys', it's worth considering how learning resources can best be organised to support any specific journey. E-learning (content) managements en e-learning environments may provide part of the solution for training materials. If these are linked by an API to the editors worksapce (in context), so much the better. On-screen tips are probably a separate issue.

Hope this helps the discussion, Mike