The best way to understand a CRM: follow one lead through it
Definitions of CRM tend to be circular — software for managing customer relationships — and leave you no wiser about what the thing actually does. So instead of defining, let us trace. Meet Priya, who runs a twelve-person interior design firm. On Tuesday at 9:40pm, a homeowner named Arjun sees her Instagram ad and taps the enquiry form. What follows is how a CRM works, step by step, with the automatic parts and the human parts clearly marked.
Step 1: capture (automatic, seconds)
The moment Arjun submits the form, the CRM creates a lead record: name, phone, requirement, and source — Instagram ad, this campaign — stamped on it. No one at the firm is awake, and no one needs to be. If Arjun had instead sent a WhatsApp message to the business number, or called and got no answer, the result would be the same: a record, created automatically. This is the first thing to understand about how CRM works — the front door is always open and nothing that walks in is lost.
Step 2: assignment and first touch (automatic, minutes)
A routing rule assigns Arjun to Sneha, the designer covering his part of the city, and pings her phone. Simultaneously, a first-touch sequence sends Arjun a WhatsApp from the business number: thanks for the enquiry, a portfolio link, and a promise of a call tomorrow morning. It is 9:41pm. Arjun, comparing three firms, has already heard back from one.
Step 3: the human conversation (manual, logged automatically)
Next morning, Sneha opens her queue. Arjun is near the top — the AI score noticed he clicked the portfolio twice within minutes, a strong engagement signal. She taps his number; the built-in dialer places the call and, when it ends, attaches the recording and her outcome tag — interested, site visit Saturday — to his record. She types nothing but the visit date, which becomes the deal's next action. The deal card moves from "new enquiry" to "site visit scheduled".
Step 4: the middle game (shared between human and machine)
Saturday's visit happens; Sneha checks in from the mobile app and voice-notes her measurements discussion. She builds the quote on the deal record and sends it over WhatsApp. Now the machine takes the watch: the CRM shows when Arjun opens the quote, and when three days pass without a reply, it nudges Sneha — or, if she is swamped, sends the polite scheduled follow-up itself and stops the instant Arjun responds. Nothing here relies on Sneha's memory, and that is the point: the follow-up that usually dies on a busy Thursday is now infrastructure.
Step 5: close, and after (automatic paperwork, human relationship)
Arjun negotiates, agrees, and the card moves to won. That stage change triggers an automation: an invoice task with the GST details ready, and a handover note to the project team with the whole conversation history attached. Six months later, another automation schedules a check-in — which is how this deal eventually produces a referral. Meanwhile Priya, the owner, never asked anyone for a status update during any of this: her pipeline board showed the deal moving the whole time.
