Skip to content
Onboarding · Database SaaS Shipped

Couchbase Day 0 · Couchbase · Capella DBaaS

Getting cloud database teams to first value in minutes, not days.

I led the redesign of Capella's Day 0 journey, from sign-up to the first successful database query, removing setup gates and guiding admins through sample data, queries and indexes.

Onboarding Funnel optimisation Time to value
Couchbase Capella welcome screen guiding a new admin to provision clusters, load sample data and execute their first query.
Couchbase Capella welcome screen guiding a new admin to provision clusters, load sample data and execute their first query.

Executive snapshot

Product
Couchbase Capella DBaaS
My role
Senior UX Designer / Manager
Users
Database administrators and cloud engineers
Definition
Day 0 = sign-up to first successful query
Method
Funnel telemetry, prototype testing, agile sprints
Partners
Product, sales engineering, cloud operations
< 5 minTime to first query
Reported outcomeFrom sign-up to first successful SQL++ query on sample data. Baseline and measurement period being confirmed.
−26%Day 1 support tickets
Reported outcomeSupport enquiries from new accounts in their first days. Baseline and measurement period being confirmed.

01 · The stakes

In database SaaS, slow setup becomes churn.

Evaluators who could not get a cluster running quickly simply left. Legacy setup asked for decisions about regions, tiers and networking before users had seen any value.

02 · My leadership contribution

Own the Day 0 experience and its success measures.

Team & collaboration model
Led design for onboarding, working with product, sales engineering and cloud operations.
Key decision
Default sensible choices and defer advanced configuration until after first value.
Stakeholder challenge
Balanced sales goals for fast evaluation with engineering concerns about defaults and cost.
Operating rhythm
Funnel reviews after each release to find and fix the next biggest drop-off.

03 · Research & insight

What we learned before we designed.

Too many decisions, too early

Six configuration gates stood between sign-up and a working database.

Networking stalled people

IP allow-lists and certificates were common drop-off points.

04 · Decision stories

The calls that shaped the product.

Each decision follows the same path: signal, insight, options, decision, trade-off, outcome.

Decision 01

Remove gates, default the rest

  1. Signal

    Funnel data showed drop-off at tier and region choices.

  2. Insight

    Evaluators wanted to try the product, not plan production.

  3. Options

    Better explanations; or remove gates with sensible defaults.

  4. Decision

    Removed six legacy gates, defaulting region and pre-loading sample buckets.

  5. Trade-off

    Defaults had to be safe on cost, agreed with cloud operations.

  6. Outcome

    First query reachable in under five minutes (reported).

05 · System view

The Day 0 funnel

A step-by-step path, each stage measured.

1 · Sign up

Minimal form, project created automatically.

2 · Cluster ready

Default region and tier, provisioned in the background.

3 · Sample data

Pre-allocated sample bucket, loaded in one click.

4 · First query

Guided SQL++ query with contextual help.

5 · Day 1 hand-off

Prompts for networking, backups and scaling.

Guided first steps

Contextual tutorials take admins from an empty account to a working query.

  • Welcome and project set-up
  • Guided sample data load
  • First query with index health tips

06 · Impact

What changed, and how we know.

Every figure is labelled with what it measures. Reported figures are being matched to their original baselines and periods.

< 5 minTime to first query
Reported outcomeFrom sign-up to first successful SQL++ query on sample data. Baseline and measurement period being confirmed.
−26%Day 1 support tickets
Reported outcomeSupport enquiries from new accounts in their first days. Baseline and measurement period being confirmed.

07 · Reflection

What I'd do differently at the next scale.

I would define the funnel metrics in a shared dashboard before the first release, not after.