A real estate lead management system, honestly scoped

Every real estate business runs its leads in a spreadsheet first, and that is the correct decision. The spreadsheet is free, everyone can read it, and it fits in one head. The problem is that it stops working gradually and then all at once, and by the time it is obviously broken, several months of leads have already been lost inside it.

This note is about where that point is and what the replacement genuinely has to do, which is less than most vendors will tell you.

The four jobs, and only four

• Capture without a human. A lead from a portal, a form or an ad arrives in the system without anyone retyping it. Retyping is where leads die.
• Assign with a clock. Every lead has an owner and a time by which first contact must happen. Both are visible to the owner's manager.
• Record what was said. Not a status dropdown. What the person actually wants, in words, attached to the lead.
• Survive a person leaving. When an agent goes, their pipeline stays legible to whoever picks it up.

That is the system. Valuation models, commission splits, portal syndication and marketing automation are all real needs and none of them are this. Adding them at the start is the most common way these projects become eighteen months long.

Where the spreadsheet actually breaks

Not at a row count. It breaks at the point where automation gets bolted onto it, which is the moment someone writes a script to pull leads in or push notifications out. That is when the failure modes stop being visible.

This has documented edges rather than folklore ones. Google's Apps Script quotas page puts script runtime at "6 min / execution". Trigger runtime totals "90 min / day" on a consumer account against "6 hr / day" on Workspace. Email recipients are "100 / day" on consumer and "1,500 / day" on Workspace. Simultaneous executions per user are capped at 30.

On the sheet side, the Sheets API limits are 300 read requests per minute per project and 60 per minute per user per project, the same again for writes, and a request that exceeds 180 seconds returns a timeout. There is no daily cap provided you stay inside the per-minute quotas.

Read those numbers as a description of a specific bad Tuesday. A brokerage on a consumer Google account runs a launch campaign, leads arrive faster than usual, the trigger that emails agents hits the 100-recipient ceiling at midday, and the leads after that point are captured in the sheet and silently never assigned to anyone. Nothing errors in a way a manager sees. The campaign simply underperforms and the ad spend gets blamed.

The number that decides whether to build

Not lead volume. Time to first contact, and how much of it you can currently prove.

Pull thirty leads from last month at random and try to answer two questions for each: when did it arrive, and when did someone first speak to the person. If you cannot answer both for most of them, that is the finding. The system you need is the one that makes those two timestamps automatic, and every other feature is negotiable against it.

If you can answer both, and the gap is small, you do not have a software problem yet and should spend the money on something else. That is a real outcome and we would rather say it than sell a build.

Build or buy

The fourth row is the one people miss. A surprising share of real estate lead problems are not missing software, they are two pieces of working software that do not speak, and the cheapest correct answer is a connector nobody will ever demo proudly. The same reasoning applies to stock and inventory, which we worked through in when a stock spreadsheet starts costing more than it saves, and the build-versus-buy arithmetic in general is in how much a custom CRM costs.

What to insist on if you do build

• Every lead carries its source, unmodified, from the moment it arrives. Reconstructing attribution later is not possible.
• First-contact time is recorded by the system, never typed by the agent who is being measured on it.
• Nothing is ever deleted. Leads are closed with a reason, and the reasons are a short fixed list somebody argued about.
• One screen a manager can open in the morning that shows only leads breaching their clock.

That last screen is the entire return on the project. If the build does not produce it in the first release, the sequencing is wrong. What real-time does and does not mean in that context we set out in a real-time dashboard for a sales team.

The honest limit

A lead management system cannot make anyone call anybody. It makes not calling visible, which is a different and smaller claim. Brokerages that install one and do not change what managers look at every morning get a tidier record of the same lost leads.

It is also worth being clear that the upstream problem is often not the system at all. Where leads arrive and nobody responds, the cause is usually staffing and incentive rather than tooling, and KF Agency's Arabic write-up of real estate lead generation for developers in Egypt (written in Arabic) covers the demand side that this system only receives.