[Meta] Kibana Onboarding

[ryankeairns]

Background

Kibana is a powerful tool that has experienced rapid growth in the number and variety of solutions offered. With each new offering, the user base becomes wider and more varied with certain segments relying upon different combinations of underlying products.

All along, the experience of configuring and using Kibana has been a hurdle for users to overcome. This recent growth has only exacerbated the issues identified in the ‘Getting started’ experience. The impact is not limited to new users as experienced users are also faced with the same new Kibana plugins, features, and enhancements.

Opportunities lie within these challenges as evidenced in the most recent Kibana All Hands Jan 2020. Following are a few relevant highlights:

Weaknesses

• Fragmented UX: “What app/language/tool do I use for my task?”
• Getting data in is difficult (80% prepping, 20% analysis)

Strengths

• Many organizations already have a lot of data in ES
• We have many of the building blocks to be competitive
• “Users typically use Elasticsearch because of its tight integration with Kibana visual analytics”

With an improved onboarding experience, we’ll be able to convert a higher rate of new users into active or power users.

Target outcomes

• Decrease the time it takes for users to get value from Kibana by helping them get their job(s) done in a timely and painless manner.
  • Requires a better understanding of user personas and their associated Jobs-to-be-Done (JTBD). What are they hiring Kibana to do? What gets in the way?
• Build upon this initial experience, encouraging users to become even more effective by learning new ways (i.e. products) to complete their jobs more efficiently.
• Provide a clear path for users to move from new, to effective, to power users.

Areas of improvement

Getting started touchpoints

• Kibana home page
• Overview pages for each solution area
• Add data section (including sample data)
• Guides + video content
• Enhanced global search + Natural Language Processing
• Product and feature tours
• What’s new

Personalized experience

• Persona-focused UX
  • Deep linking from Cloud into specific solution Overview pages
  • Tailor the experience to their role or job - “Welcome to Kibana, what would you like to do?”
• How do Elastic Package Manager (EPM), Add data, and sample data support this experience?
• How do overview pages, guides, docs, website content, etc., support this?
• How do we continue to provide contextual guidance beyond the ‘getting started’ experience? (e.g. tours of new features, intelligent recommendations based upon telemetry, etc.)

Research

We have educated guesses based upon anecdotal feedback that gives a good idea that a) there are in fact issues and b) where those challenges likely reside. By conducting a little more research, we can pinpoint problem areas and formulate a targeted approach.

Does the home page need some help? Of course.
Could we make some guesses that improve it? Sure.
How much better could it be if we tailored the experience? Let’s find out.

What we know

• Kibana has a steep learning curve, but it’s getting easier with things like Lens
  • In particular, getting data into the Kibana UI is difficult
  • Index patterns present an immediate hurdle
  • The first user in a new instance gets all the tips and empty states, subsequent users may never see these (lack of personalization)
• There are new products all the time, users aren’t clear on what they do or how to use them
• Setting up and learning new products (even for experienced users) can be challenging
• Users are here to do a job, we could do more to streamline their workflows
• Traditionally used by:
  • Admin, technical, DevOps audience
  • Elasticsearch is the main gateway, then Kibana follows for visualizing ES data (e.g. discover, visualize, logs, dashboard, etc.)
• We have a growing emphasis on Cloud which is becoming THE entry point
• Applications are beginning to crossover more as services are being built to better support these workflows (URL service, embeddables, etc.)

What we don’t know

• All the types of users operating within Kibana
  • Security/APM/Geo/Biz/Data analysts and more
• Precisely how different users use Kibana - how do they flow through the application(s) for a given job?
• In terms of JTBD,, what has each user type hired Kibana to do?

How we can learn more

• Surveys
• Customer interviews
• Internal interviews (product managers, engineers, internal users, consulting, support, etc.)
• Field studies

Through these activities, we can gain a better understanding of...

• how jobs map to personas, solutions, products,
• what are the most challenging aspects of learning Kibana, and
• which patterns work well and we should do more of.

Roadmap

This will be a long-running effort, spanning numerous smaller projects, delivered in an iterative fashion. While the Core UI team does not ‘own’ all of these projects, it can drive the effort and work with solution teams as the experiences grow deeper and cross internal boundaries..

[ ] denotes projects not managed by Core UI

Sub-projects

• Elastic Console home page - [Home] Redesign Kibana Homepage to reduce clutter #25734
• Kibana solution landing page - [Onboarding] Kibana landing page #61549
• Elastic Console "Choose your solution" log in page - Display a soultion-oriented "Choose your own journey" interstitial page #71737
• Getting started / [solution] / [product] tours - Add Product Tour to Kibana #18787 Build tweaks #505 (EuiTour)
• Add data / Sample data / [EPM]
• Guides - (needs issue; coupled with Home page), Kibana Guides for Cloud users #35717 (Cloud guides)
• [Solution-based overview pages]
• [Personalization]
• Enhanced global search + NLP - [Meta] Global Search #57576
• [Index pattern changes]

Next steps

• 1. Conduct initial interviews
• 2. Compile and share insights
• 3. Review and update home page designs from last year
• 4. Pilot EuiTour from landing page
• 5. Plan next steps (for sub-projects) based upon findings

[elasticmachine]

Pinging @elastic/kibana-design (Team:Kibana-Design)

[ryankeairns]

Closed in favor of the cross-team Getting Started project.