Engagement & Lead Score: engagement timeline and configurable scoring

Conversa Labs

Conversa Labs

Last updated on Jun 27, 2026

Overview

Engagement & Lead Score measures how engaged each contact is and computes a configurable lead score. Every relevant interaction becomes an event on a timeline tied to the contact β€” and that timeline survives even if the originating conversation or deal is deleted, because the events are stored independently.

You define an event catalog (a kind of behavior "pixel"), organize the events into point groups and assign weights to each one. Real product events β€” messages, deals, follow-ups, payments, bookings, catalog sales, emails β€” are mapped to that catalog and add (or don't add) points according to the weight. With this, your team prioritizes the hottest leads.

This module is the lead axis (the contact), complementary to the Sales module, which is the actor axis (the rep). When it makes sense, the same event can grant dual credit: engagement points for the contact and, if the actor is a rep, XP for the rep in the Sales module.

Prerequisites

  • The Engagement & Lead Score module enabled for the account (optional feature, turned on by an administrator/operator).
  • The engagement management permission to configure the event catalog, groups and weights.
  • Real events flowing through the platform (conversations, messages, CRM, payments, calendar, follow-ups, catalog) to feed the scoring.
  • For the rep's dual XP credit, the Sales & Gamification module must also be enabled.

Step by step

  1. Build the event catalog. Register the events you want to track (each with a name, group, icon and default points).
  2. Define the point groups. Group events by theme (for example, support, purchase, marketing).
  3. Adjust the weights. Give more points to the most important events; use weight 0 to just record (track) without scoring.
  4. Choose the dedup policy per event: count always, once per contact, or with an interval (cooldown).
  5. Track the timeline. On the contact panel, see the engagement timeline and the lead score.
  6. Prioritize and segment. Sort lists by score and build segments of the hottest leads.
  7. (Optional) Track external events. Use the tracker (pixel) with a token to record events outside the platform.

Settings & options

  • Event catalog (definitions): key, name, group, icon, default points and whether it's active.
  • Point groups: named counters that add points by theme.
  • Weights: score per event (including 0 to record without scoring).
  • Deduplication: count always, once per contact, or with a cooldown (configurable interval).
  • Lead score: overall and per-group score, shown on the contact and usable to sort/segment.
  • Dual credit: the same event can score the contact (engagement) and the rep (XP in Sales).
  • Tracker (pixel): a public token-verified endpoint for external events.
  • Retention: an optional retention period for the timeline events (no point decay in v1).

Use cases

  • Prioritize the lead queue by the contacts with the highest score.
  • Create a "hot leads" segment for a targeted sales action.
  • Track a contact's journey on a single timeline, even after archiving conversations.
  • Record behaviors outside the platform (a page visit, an email open) via the pixel.

Tips, limits & best practices

  • Start with a few high-value events; refine the weights over time.
  • Use weight 0 to trial a new event before deciding how much it's worth.
  • Pick the right dedup policy to avoid inflating the score (e.g., "once per contact" for sign-ups).
  • Remember: in v1 there's no point decay β€” the score is cumulative.

Troubleshooting

  • The lead score doesn't change: confirm the event is active in the catalog, with weight greater than 0, and that real events are happening.
  • An event counted too much: review the dedup policy (you may need "once per contact" or a cooldown).
  • The timeline disappeared after deleting a conversation: it doesn't disappear β€” engagement events are preserved independently of the source.
  • The external pixel isn't recording: check the tracker token and the request limits.

See also