Every business has records its CRM never heard of. An agency has projects and retainers. A clinic has appointments and treatment plans. A property firm has listings, viewings and handovers. Forcing all of that into "contacts" and "deals" is how CRMs end up half-abandoned, shadowed by the spreadsheets that hold the real data. The data engine in Autometa CRM exists so the system can hold the shape of your actual business — this post tours how.
Custom objects with real field types
Custom objects let you define the records your operation actually runs on, with more than twelve field types that do real work: formula fields that compute margin from value and cost, rollups that sum every payment on an account up to the account itself, lookup relations that link a viewing to its listing and its buyer, plus currencies, multi-selects and auto-generated IDs for clean referencing.
Formulas and rollups matter more than they sound. They move the arithmetic your team currently does in exported spreadsheets into the record layer itself — always current, visible to automations and AI, and impossible to forget to update.
Pipelines with judgment built in
Deal pipelines are drag-and-drop Kanban with custom stages — table stakes — but the useful parts are the opinions layered on top. Stage win probabilities turn an inflated pipeline into a weighted forecast automatically. Deal ROT alerts flag records that have sat in a stage past their expected shelf life, before the quarter ends rather than after. The board stops being a place deals are displayed and becomes a place stalls are caught.
Dashboards that answer, not decorate
The dashboard layer ships with eight-plus chart types aimed at questions managers actually ask: conversion funnels to find the leaking stage, pipeline velocity to see whether this month is genuinely slower, rep leaderboards for coverage and coaching, cohort trends to learn whether this quarter’s leads behave like last quarter’s. Every chart reads live records — including your custom objects and formula fields — so "can we get a report on that" stops being a data-export project.
Booking pages and the calendar dance
Public booking pages end the "does Tuesday work?" thread: a prospect picks a slot, round-robin scheduling distributes it fairly across the team, and a Google Meet or Zoom link lands on both calendars automatically. Because the booking is a CRM event, not a third-party calendar entry, it attaches to the lead’s record and can trigger workflows — a reminder sequence, a pre-call research task, a no-show follow-up.
Work that lives next to the record
The engine also carries the work around the records. Tasks and project sprints run on Kanban boards with backlogs, subtask checklists and dependency tracking — enough project management for delivery teams without buying another tool. Rich-text notes support @mentions and version history, and two-way Google Drive and Dropbox sync attaches the proposal deck to the deal without re-uploading it anywhere.
Getting data in without a services engagement
A data engine is only useful if your data reaches it. The public form builder creates embeddable multi-step forms with conditional logic that write straight into CRM records — lead capture with no middleware. For existing data, the import wizard takes CSV and JSON with automatic field mapping, duplicate detection and tag assignment, so migrating from spreadsheets or another CRM is an afternoon, not a project plan.
The through-line is altitude: model the business once, at the record layer, and everything above it — pipelines, dashboards, automations, AI — inherits the model for free. That is what a system of record is supposed to mean.