Skip to content

Sarva Personalization

In development

Turn customer behaviour into the next best experience.

Behaviour becomes a profile, a profile becomes a segment, a segment becomes a decision — while the visit is still happening. Sarva Personalization is in development.

How a decision is made

Five steps, measured in milliseconds.

Personalisation that runs on last night's segment export is just a slower version of the same page. The value is in doing it during the session.

  1. Behaviour0ms

    A visitor does something — views a page, opens an email, adds to a cart.

  2. Profile~5ms

    The event is matched to a known customer, or held as an anonymous identity.

  3. Segment~15ms

    Membership is recalculated against real-time audience definitions.

  4. Decision~25ms

    Rules and scores select what should happen next for this specific person.

  5. Experience~40ms

    The chosen experience is delivered and logged for reporting.

Sarva Personalization/decisions/trade-pricing
  1. Behaviour
    Viewed 3 bearing SKUs, opened pricing twicelive session
  2. Profile
    Matched to acc_8841 · trade customeridentity resolved
  3. Segment
    High intent · no open quotation1 of 3 audiences
  4. Decision
    Show trade pricing and request-a-quotevariant B
  5. Experience
    Rendered in 40ms · logged for reportingdelivered
Planned capabilities

What Sarva Personalization is designed to do.

Audiences

  • Real-time audiences
  • Anonymous identities
  • Membership rules
  • Suppression

Scoring

  • Intent scoring
  • Engagement scoring
  • Propensity signals
  • Decay rules

Decisioning

  • Decision tables
  • Next-best action
  • Recommendations
  • Fallbacks

Delivery

  • Website personalization
  • Campaign targeting
  • Behavioural triggers
  • A/B testing
Use cases

What people use it for

Ecommerce lead

Every visitor sees the same homepage regardless of history.

Returning trade customers see their pricing and reorder options immediately.

Demand generation

High-intent visitors leave without ever speaking to sales.

Intent crossing a threshold changes the page and notifies the account owner.

Growth

Testing requires a separate tool with its own definition of a user.

Experiments run against the same profiles the rest of the suite uses.

Where this stands today

Sarva Personalization is in development. It depends on Sarva Data, which is in early access — real-time decisioning is only meaningful once identity is resolved.

Next step

Start with Sarva Personalization. Add the rest when you need it.

Tell us what your team is doing manually today and we will show you where in the suite it belongs.