Skip to main content
CRM Integration

AI Lead Scoring for Salesforce

Your objects, your automations, your fields. Trailspark adds Hot, Warm, or Cold scores with written reasoning on Leads and Contacts, plus propensity and buying-group coverage on the Account.

What syncs between Trailspark and Salesforce

Trailspark reads the Salesforce fields you map and your opportunities to score each account, then writes results to custom fields you create, updated after each evaluation.

Salesforce
Trailspark

Salesforce sends to Trailspark

  • From Salesforce

    Account, Lead & Contact fields

    Industry, employee count, market segment, region, and job title from the fields you map, used for ICP matching and role-based scoring.

  • From Salesforce

    Opportunities

    Synced onto accounts so deal history informs scoring. On by default; limit it to the record types and opportunity types you choose.

  • From Salesforce

    CRM Role Source Field

    Point Trailspark at a Contact field where your team records buying roles, and those assignments are treated as authoritative.

Trailspark writes to Salesforce

  • To Salesforce

    Trailspark Score

    Hot, Warm, or Cold, written to a field you create on Lead and Contact. Updated after each evaluation.

  • To Salesforce

    Trailspark Reasoning

    The plain-language reason behind the score, in a Long Text Area field reps can read before a call.

  • To Salesforce

    Evaluation Date, ICP & Score Status

    When the evaluation ran, which ICP scored the record, and whether that score is current, retired, or not evaluated, so Flows can tell a fresh score from a stale one. ICP and status are optional.

  • To Salesforce

    Account propensity

    High, medium, or low on the Account, with its ICP and score status. Unconverted Leads can take propensity as a fallback. All optional, mapped per field.

  • To Salesforce

    Buying-group roles & coverage

    With buying-group push on: each person's roles on Lead and Contact, and Account fields for stage, engaged and missing roles, and people still needed. Only mapped fields are written.

Use cases

Prioritize inbound with the reason attached

Trailspark scores each person against your ICP using product usage, enrichment, and the Salesforce data you map, then writes Hot, Warm, or Cold with written reasoning to the Lead or Contact. Your Flows route on the score, and reps read the reason before they call.

Create records only for people worth a rep's time

When someone without a Salesforce record reaches the score levels you choose, Trailspark creates one: a Contact on the matching Account, or a Lead with company details filled in. Everyone else stays in Trailspark, out of your CRM.

See buying-group coverage on the Account

Buying Groups tracks who fills each role at an account. Push it to Salesforce and the Account shows its stage, engaged and missing roles, and how many people are still needed, next to its propensity.

Setup

Your Salesforce admin's part takes about five minutes. The rest happens in Trailspark.

  1. 1

    Your admin creates a Connected App

    A one-time step in Salesforce. OAuth with PKCE means no Client Secret is stored and there's no managed package to install.

  2. 2

    Map your fields

    Trailspark loads your Salesforce schema when you connect. Map the fields it reads and the custom fields that receive scores.

  3. 3

    Choose what gets written

    Pick the score levels that can create new records and whether buying-group fields are pushed. Everything else stays as is.

Connect Salesforce with Trailspark step by step

Our docs cover creating the Connected App, connecting a production or sandbox org, field mapping after you connect, opportunity sync, and troubleshooting.

Read the Salesforce setup guide (opens in a new tab)

How the Salesforce integration works: common questions

A Connected App, plus custom fields on Lead and Contact for Score (text), Evaluation Date (DateTime), and Reasoning (Long Text Area). ICP, Score Status, account propensity, and buying-group fields are optional. Trailspark writes only to the fields you map.

Trailspark matches people by email. When the person's account is linked to a Salesforce Account it looks for a Contact first, otherwise a Lead, and falls back to the other object. Lead and Contact fields are mapped separately.

No. Trailspark writes only to the custom fields you map, so Einstein and the automations built on it keep running. You can run both side by side and compare before deciding what to keep.

System Administrator access to create a Connected App, which takes about five minutes. You paste its Consumer Key into Trailspark as the Client ID; with PKCE, no Client Secret is needed. Production and sandbox orgs are both supported.

Yes. Opportunity sync is on by default once Salesforce is your active CRM, and attaches opportunities to accounts to inform scoring. You can limit it to specific record types and opportunity types, or turn it off.

No. Turn on buying-group push and map only the fields you want. Unmapped fields are never written.

Trailspark is billed on the unique accounts it evaluates each month, not seats or CRM contacts. Every plan is the full platform, Free included; see pricing for the tiers, or talk to us about custom volume.

Other integrations and pages

HubSpot

CRM · Two-way sync

RudderStack

Marketing and CDP · Evidence in

Buying Group Intelligence

Buying roles and gaps, written to your CRM

See Trailspark and Salesforce on your own data.

Start free. Connect Salesforce with OAuth. Watch scores land in custom fields you create on the records your reps already use.