Analytics Dashboard

This portfolio is built in public — that includes what we track. Here's a live view of every event we collect. Learn more about our data policy ↓

Tracked events 7

Prompt Chip Clicks

Fired when a visitor selects one of the predefined conversation starters.

What's your most challenging project?
44
What does 0→1 product design look like?
8
How do you approach user research?
5
Tell me about your AI product experience
5
How do you work with engineering teams?
4
What's your design leadership style?
2

Chat Messages

Fired on every message sent to the AI. Custom messages are redacted if content moderation flags them.

78Total messages

83%From prompt chips

17%Custom questions

41Avg custom message length (chars)

AI Content References

Fired mid-stream when Claude surfaces a project, blog post, or design snippet.

robin-ai
30
draftwise
6
first-designer-startup
3
total-platform
2
analytics
2
design-methodology
1

Side Panel Clicks

Fired when a visitor clicks a referenced content card in the side panel.

robin-ai
45
draftwise
7
design-methodology
2
first-designer-startup
2
total-platform
1
analytics
1
cross-collaboration
1

Resource Views

Fired when a project case study or blog post page is loaded.

robin-ai
128
total-platform
29
draftwise
16
accessibility
12
acceptance-criteria
10
portfolio-for-all
10
design-methodology
7
design-management
6
cross-collaboration
6
first-designer-startup
6

Page Views

Fired on every page navigation.

Path	Views	Share
/	129	38%
/project/robin-ai	47	14%
/brain	43	13%
/blog	29	8%
/work	21	6%
/updates	13	4%
/project/total-platform	11	3%
/analytics	10	3%
/blog/draftwise	5	1%
/blog/acceptance-criteria	5	1%
/project/portfolio-for-all	5	1%
/blog/accessibility	4	1%
/blog/first-designer-startup	4	1%
/blog/cross-collaboration	4	1%
/blog/design-methodology	3	1%
/blog/design-management	3	1%
/blog/technical-knowledge	3	1%
/blog/working-in-design	2	1%
/blog/go-to-market	1	0%

Safeguarding: Flagged Messages

Safeguarding

A subset of chat_message_sent events where content moderation triggered. Message text is always [REDACTED]. Tracked so we can monitor misuse without storing the content publicly.

0flagged messages this period

Data Policy

Every event is documented above. Expand a section to read the full policy.

Data & Privacy Policy

This portfolio is built in public — that includes being transparent about what data we collect and why.

What we collect

All analytics are powered by PostHog, an open-source analytics platform. We collect seven event types (listed above), all documented in this file and visible in the live dashboard. Every event captured is either interaction data (what you clicked, what pages you visited) or AI conversation metadata (how long your messages are, which conversation starters you used).

What we do NOT collect

No names, email addresses, or personal identifiers. No IP addresses stored in our analytics (PostHog anonymises these). No payment or contact information — this site has none. No cross-site tracking, no third-party ad networks, and no session recordings or heatmaps.

Content moderation and safeguarding

Messages sent to the AI chat are scanned for severe profanity and prompt injection attempts before the analytics event is sent. If flagged, the actual message text is never stored in our analytics — the event records message_text as "[REDACTED]" and a count of flagged events is shown in the dashboard so we can monitor for misuse. The message is still sent to Claude, which handles inappropriate content gracefully on its own.

Cookie consent

Analytics only initialise after you accept cookies via the consent banner. If you choose "Continue without", PostHog is never initialised — no cookies are set, no events are captured. You can reset your choice at any time by clearing localStorage for this domain.

Data retention

PostHog retains event data for 1 year on the free plan. We do not have custom retention settings configured.

Maintainer note

This file is the source of truth for analytics on this site. Every time a new PostHog event is added to the codebase, add it here under events in the frontmatter. The analytics dashboard reads this file and automatically renders a widget for each event with dashboard: true.