Homepage 'How It Works' inverts journey order — drops the ingestion step, says datasets are selected during use-case definition

[saadqbal]

Bug

The homepage overview/tracebloc.mdx presents a 5-step "How It Works" that omits data ingestion entirely and frames "Select datasets" as something that happens during use-case definition (Step 2). This contradicts the deeper workflow docs, which correctly position ingestion as a prerequisite for use-case creation.

The contradiction sends new users into the platform with the wrong mental model: they create an account, click "Define a use case," look for a dataset selector, and find nothing because they haven't ingested any data yet.

Audit across the docs site

Page	Order it presents	Correct?
Homepage /overview/tracebloc — "How It Works"	Install → Define use case (select datasets) → Invite → Build → Compare	❌
Section overview /create-use-case	Prereqs (client deployed, dataset cleaned) → Prepare & ingest → Define → Set metrics → Evaluate	✅
/create-use-case/prepare-dataset	Ingest → then "Define Use Case"	✅

Offending source

overview/tracebloc.mdx:63-65:

<Step title="Define a use case">
  Select datasets, set evaluation metrics. Use a template or build your own.
</Step>

Same wording is repeated in the "Next Steps" card at overview/tracebloc.mdx:104-106:

<Card title="Create a Use Case" icon="plus" href="/create-use-case/prerequisites">
  Select datasets, define metrics, invite contributors.
</Card>

Why this matters

• Homepage traffic is much higher than the deep workflow pages — most newcomers form their mental model here.
• The internal contradiction (homepage vs. /create-use-case) is itself confusing — a careful reader can't tell which page is wrong.
• Corresponds to the same bug class fixed in tracebloc/data-ingestors#65 and #66 (Readme prereq listed "active use case" as required before ingest).

Suggested fix

Add an explicit ingestion step to the homepage and reword Step 2 so it doesn't read as "datasets are configured during use-case creation." One option:

<Step title="Create your AI workspace">…</Step>
<Step title="Bring your data">
  Run the ingestor to validate, preprocess, and stage your dataset on your cluster. Only metadata syncs to the web app.
</Step>
<Step title="Define a use case">
  Pick from your ingested datasets, set evaluation metrics. Use a template or build your own.
</Step>
<Step title="Invite anyone">…</Step>
<Step title="Build together">…</Step>
<Step title="Compare and decide">…</Step>

Apply the same rewording to the "Next Steps" card.

Acceptance

• Homepage's "How It Works" reflects the same order as /create-use-case: ingestion is a step before use-case definition, not subsumed into it.
• "Select datasets" no longer reads as a UI action performed during use-case creation; it reads as picking from previously-ingested datasets.
• Homepage and /create-use-case no longer contradict each other.