LoginSign Up
← Back to all sparks
Holistics logo

Holistics

ANALYTICS
Velocity8.8

Self-service BI and data analytics platform

www.holistics.io ↗

Holistics builds AI governance and docs-as-analytics in parallel, shipping both weekly

analyticsbiai-governanceself-service-analyticsdata-documentation
◆Current state
Holistics is in an active multi-track build across three distinct areas: AI governance (trace AI-triggered queries, per-user MCP access, data redaction for AI), self-service intelligence (anomaly detection, root-cause analysis), and a UX overhaul (new universal sidebar, dark mode). The pace is 3–5 releases per week with a high proportion of user-visible changes.
◆Where it's heading
Two directions are clearly compounding. First, AI governance — each release adds a new layer of admin control or observability over how AI agents access data, positioning Holistics for enterprise buyers who need to audit AI behavior in the warehouse. Second, Markdown as data catalog — integrating documentation directly with models and dashboards points toward a data layer where docs live in code rather than a separate UI. The navigation overhaul clears the UX runway for both.
◆Prediction
Holistics will continue deepening AI observability (attribution, access controls, and data redaction all shipped within weeks suggest a coordinated feature set) and extend the Markdown-as-data-catalog concept toward lineage visualization or schema documentation enforcement.

◆Recent moves

  1. 13h ago

    Universal sidebar replaces header-based navigation across the whole app

    A universal sidebar replaces the old header-and-panel navigation, putting dashboards, data, and tools in one place. This is infrastructure for a growing feature surface — the old navigation was becoming a bottleneck as AI and data-catalog features accumulated.

    View source ↗
  2. 2d ago

    🔍 Trace AI-triggered queries and jobs

    ⚡ SPARK

    AI-triggered query attribution is now surfaced in the Jobs API and warehouse query comments — teams can see whether a query came from AI, which surface triggered it, and how the caller authenticated. This extends Holistics's AI governance capability surface meaningfully.

    View source ↗
  3. 2d ago

    Dashboard theme follows system light/dark preference

    Dashboard themes now follow system light/dark preference, completing the dark-mode rollout that started earlier. A visible UX quality signal for teams embedding Holistics dashboards in products with their own theme systems.

    View source ↗
  4. 6d ago

    Canvas Dashboard requires explicit interaction list; no implicit connections assumed

    Canvas dashboards now require every active interaction to be listed explicitly in an explicit_interactions block — no more implicit connections based on field overlap. This reduces debugging time for complex dashboards and aligns with the code-as-configuration pattern Holistics uses for the rest of its data layer.

    View source ↗
  5. 8d ago

    Markdown docs become first-class objects linked to models, datasets, and dashboards

    ⚡ SPARK

    Markdown files in a Holistics project now render frontmatter as a Properties panel and allow links to models, datasets, and dashboards — making documentation a node in the analytics graph rather than a disconnected text file. This is the first step toward a living data catalog built in code.

    View source ↗
  6. 9d ago

    Per-user MCP access control, separate from in-app AI access

    Admins can now grant or revoke MCP access independently from in-app AI access, per user or group. This gives enterprises a finer control plane as Holistics expands its AI surface — a governance feature that directly enables the AI tracing capability shipped the same week.

    View source ↗