What's really going on

You do about 90 jobs a year, and roughly 72 of them are people you already know. That's not a pipeline. That's a rolodex with a service history. The real risk in your business isn't losing leads — it's forgetting to call a repeat client when their next cleaning is due, or losing the notes on what their last job cost and what their building needed.

What to do instead

1. Google Sheets (free) — one row per client: name, site, last service date, next-due date, price, notes. Four office people can edit it live. That's your "CRM."
2. Jobber (check current pricing) — if you want scheduling, quotes, and invoices in one place for the field crew, this is built for duct/cleaning trades. Only worth it if invoicing and dispatch are the pain, not client memory.
3. Habit: every Friday, one person opens the sheet and calls anyone whose next-due date is inside the next 30 days.

First action, today: make the sheet. Columns across the top, paste in your last 12 months of jobs from your invoices, sort by client. One hour, one person.

What you're being oversold

Someone probably pointed you at Pipedrive or HubSpot. Both are wrong here. Pipedrive is a sales pipeline — you don't have a sales pipeline, you have a repeat book. HubSpot Free looks tempting but it's built to funnel you into paid Sales Hub, and you'd be paying to track "leads" you already have on speed dial. Doing nothing costs you the repeat calls you forget to make — a handful of jobs a year that quietly go to whoever remembered first.

When to revisit this

When you're past 25 jobs a month, or you hire a dedicated salesperson chasing new buildings.

The part worth getting right

The one call is this: is your real bottleneck remembering clients or running the jobs? If it's memory — who's due, what you charged, what the building needs — a sheet solves it this week and you're done. If it's the job itself — quotes going out slow, invoices chased twice, techs showing up to the wrong site — then Jobber earns its price and a sheet won't. Pick wrong and you either pay monthly for software your team won't open, or you keep losing an afternoon a week to admin that a $50/month tool would kill.