CRM automation for leads: how we booked 40 qualified calls a month for a Mississauga mortgage agent

Content brings people in. If there is no system to capture them, most of those people are lost. A viewer watches your Reel, taps your profile, looks around, and scrolls on. Two weeks later they cannot remember your name.

The keyword CTA system

Every Reel ends with a single keyword CTA. Comment BUDGET and I will send you the free calculator. When a viewer comments that keyword, a GoHighLevel workflow fires immediately. The workflow sends a DM within 30 seconds with the promised resource. That DM also contains one qualifying question: Are you looking to buy in the next 3 to 6 months? The answer segments the lead automatically.

The nurture sequence

Leads not ready to buy in the next 3 months go into a longer nurture sequence: one educational message per week, each referencing a recent Reel the agent posted. Leads buying in the next 3 months go into a high-touch sequence: a human-sounding follow-up message, a calendar link, and a reminder 24 hours before the booked call.

What changed for the client

Month one after automation went live: 42 leads captured, 11 consultations booked, 3 mortgages closed with a direct attribution trail back to specific Reels. The client could now see exactly which content was generating revenue. The content strategy shifted immediately: formats that generated the most comment keywords got more posting frequency.

The volume at which manual follow-up breaks

Here is the number that decides whether you need any of this. In August 2026 our ads dashboard recorded 469 leads across six client accounts in a single month. That is roughly 78 leads per client, or two to three arriving every day, on top of whatever the organic content brings in through DMs.

Two or three a day sounds manageable until you try it while showing houses. The follow-up window that matters is the first hour, and an agent in a car cannot reliably answer a form fill from ninety minutes ago. That gap is where most paid budget quietly dies: the lead was real, the ad worked, and nobody called back in time.

A client who hit that wall

Manish Puri, an Oakville realtor we worked with from June 2025 to June 2026, called us partway through the engagement because the inbound had outgrown him. His account had gone from nothing to 5.88 million views, and the volume of enquiries coming off it was more than he could work through between appointments. He asked us to build something that would hold the line while he was busy. That is the actual origin of most CRM work we do: not a strategy, a bottleneck.

What gets automated, and what does not

The parts worth automating are the mechanical ones. A keyword reply in the comments, an instant first message so the lead knows they reached a real business, a tagged record in the pipeline so nothing sits unassigned, and a reminder sequence for the follow-ups that would otherwise be forgotten on day three and day nine.

The part that should never be automated is the conversation. A lead who has replied twice and asked about a specific street does not want a sequence, they want the agent. The point of the automation is to buy that agent enough time to get there, and to make sure the ones who would have been dropped are still warm when they do.

Measured properly, the win is rarely more leads. It is the same leads, reached sooner, with fewer falling through the gap between the form fill and the phone call.

What to measure, and what automation cannot fix

Three numbers tell you whether any of this is working. Speed to first contact, measured in minutes rather than hours. The share of leads that got a human reply at all, which is usually far lower than agents expect. And the share that turned into a booked appointment, which is the only one that pays for anything.

Lead count is the number most people report and the least useful of the four. Four hundred and sixty-nine leads in a month is a good headline and tells you nothing about whether anyone spoke to them.

The limit worth being honest about

Automation makes a good lead source better and a bad one cheaper to be disappointed by. On 19 August 2026 our own dashboard flagged six of ten live campaigns as costing too much per lead. No sequence fixes that. A CA$34.50 lead that gets an instant reply is still a CA$34.50 lead, and the work to bring that number down happens in the creative, not the CRM.

The order matters. Get the content earning attention, get the cost per lead into a range you can live with, then automate the follow-up so the leads you already paid for stop leaking. Building the automation first is a common and expensive way to industrialise a process that was not working yet.

One more thing worth setting up before any of the sequences: a single place where every lead lands, whatever door it came through. Comment keywords, DMs, form fills and phone enquiries tend to accumulate in four different inboxes, and a lead that exists in only one of them is a lead nobody owns. Most of the value people attribute to automation is really just the effect of having one list instead of four.