# AI Lead Scoring for Airtable | Trailspark > Keep running your pipeline in Airtable. Trailspark scores each account and writes Hot, Warm, or Cold with written reasoning to fields you create. _URL: https://www.trailspark.ai/integrations/airtable_ --- CRM Integration # AI Lead Scoring for Airtable Your base, your tables, your field names. Trailspark reads the tables you map and writes Hot, Warm, or Cold with written reasoning back to fields you create, plus account propensity and buying-group coverage on your Companies table. [Sign up free](https://app.trailspark.ai/register)[Setup guide (opens in a new tab)](https://docs.trailspark.ai/docs/connecting-airtable)[About Airtable (opens in a new tab)](https://www.airtable.com) ## What syncs between Trailspark and Airtable You map your Contacts, Companies, and optionally Deals tables once. Trailspark reads the fields you choose to score each account, then matches people by email and writes results to the fields you point it at. ![Airtable](https://www.trailspark.ai/images/integrations/airtable-logo.svg) ![Trailspark](https://www.trailspark.ai/images/logos/trailspark-ai-logo-color.svg) ![Airtable](https://www.trailspark.ai/images/integrations/airtable-logo.svg) Airtable sends to Trailspark - From Airtable ### Contact & Company fields Industry, employee count, region, job title, and anything else you map, used for ICP matching and role-based scoring. Trailspark follows your Contact-to-Company link field to reach the account. - From Airtable ### Deals Optional. Map a Deals table and closed-won history, close dates, amounts, and stage inform scoring. Your won indicator can be a checkbox, a date, or the values you name. Trailspark writes to Airtable - To Airtable ### TrailSpark Score Hot, Warm, or Cold, written to a field on your Contacts table. Single line text, or a single select with your own options. - To Airtable ### TrailSpark Reasoning The plain-language reason behind the score, written to a Long text field your team can read in the record, a grid view, or an interface. - To Airtable ### Evaluation Date, ICP & Score Status When the evaluation ran, which ICP scored the record, and whether the score is current, retired, or not evaluated, so an automation can tell a fresh score from a stale one. ICP and status are optional. - To Airtable ### Account propensity & reasoning High, medium, or low on the Companies table, with its ICP, score status, and the explanation behind it. All optional, each mapped to a field you choose. - To Airtable ### Buying-group coverage With buying-group push on: stage, engaged and missing roles, engagement and known completeness, and how many people are still needed, written to your Companies table. Per-contact roles are a Salesforce and HubSpot feature. ![Trailspark](https://www.trailspark.ai/images/logos/trailspark-ai-logo-color.svg) ## Use cases ### Use Airtable as a CRM, and score the pipeline in it Trailspark treats your mapped tables [the way it treats Salesforce or HubSpot](https://docs.trailspark.ai/docs/crm-integration-overview): your records are read for ICP matching, scored against your product, marketing, and enrichment data, and the results come back to your base. No second CRM to buy or migrate into. ### See the whole buying group on the account Your Companies table can carry the [buying group's stage](https://docs.trailspark.ai/docs/buying-groups-overview), which roles are engaged, which are still missing, and how many more people it takes to complete the group, next to the account's propensity score. ### Let your views and automations act on the score Once the score, reasoning, and evaluation date are fields in your own base, they work like every other field: filter a view to Hot accounts, group by propensity, or trigger an Airtable automation when a score changes. ## Setup Two stages: authenticate, then map your tables. Field Mapping and Destinations open once table mapping is done. 1. 1 ### Connect your base Authorize with OAuth, or paste a Personal Access Token if your organization restricts OAuth connections. 2. 2 ### Map your tables Tell Trailspark which table holds contacts and which holds companies, and add your Deals table if you want deal history in the score. 3. 3 ### Choose what gets written Create fields in your base for the score, reasoning, and dates, then pick them from a dropdown. Nothing is written to a field you haven't mapped. ### Connect Airtable with Trailspark step by step Our docs cover both ways to authenticate, the table mapping wizard, the deal fields Trailspark can use, the fields to create for results, and troubleshooting. [Read the Airtable setup guide (opens in a new tab)](https://docs.trailspark.ai/docs/connecting-airtable) ## Common questions ### Yes. Once your tables are mapped, Trailspark works with your base the way it works with Salesforce or HubSpot: your Contacts and Companies tables are read for ICP matching and scoring, a mapped Deals table lets deal history inform the score, and the results come back to fields you create. There is no second CRM to buy or migrate into. ### No. Your base stays where it is and your team keeps working in it. Trailspark reads the tables and fields you map, and writes results to fields you create in that same base. Nothing is migrated, and no structure is imposed on your tables. ### You tell it during setup. Airtable's schema is open, so after you authenticate, [a mapping wizard](https://docs.trailspark.ai/docs/connecting-airtable) asks which table holds contacts and which holds companies, both required, and lets you add a Deals table. A linked field connecting each contact to their company is recommended, and is what makes company-level fields work. ### Either. OAuth is the recommended path and refreshes its own tokens. If your organization restricts OAuth connections, create a [Personal Access Token](https://docs.trailspark.ai/docs/connecting-airtable) with the data.records:read and schema.bases:read scopes, grant it your CRM base, and paste it into Trailspark. A token has to be regenerated by hand if it expires or is revoked. ### [One CRM is active at a time](https://docs.trailspark.ai/docs/crm-integration-overview). Connecting Airtable deactivates the CRM that was connected before it, and the other way around. Your marketing, product, and enrichment sources are unaffected and keep feeding scoring. ### Only if you ask it to. Record creation is off by default: Trailspark matches a person by email in your Contacts table and updates that record. Turn creation on and you choose which score levels are worth a new Contact record, so a cold evaluation doesn't fill your base. ### The ones you pick from a dropdown, under whatever names you already use. On Contacts: the score, the evaluation date, the reasoning, and optionally the ICP and score status. On Companies: propensity, its ICP, score status, the account's reasoning, and [buying-group coverage](https://docs.trailspark.ai/docs/airtable-destination). Use a Long text field for reasoning; a field too small to hold it rejects the whole update. ### 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](https://www.trailspark.ai/pricing) for the tiers, or [talk to us](https://www.trailspark.ai/contact) about custom volume. ## Other integrations and pages [Integration ### Salesforce CRM](https://www.trailspark.ai/integrations/salesforce)[Integration ### HubSpot CRM](https://www.trailspark.ai/integrations/hubspot)[Integration ### Segment Customer Data Platform](https://www.trailspark.ai/integrations/segment)[Product ### How scoring works The full product page.](https://www.trailspark.ai/ai-lead-scoring) ## See Trailspark and Airtable on your own data Start free. Connect Airtable with OAuth. Watch scores land in custom fields you create on the records your reps already use. [Sign up free](https://app.trailspark.ai/register)