Everyone in Your Database Sells Eventually

1. The Asset Nobody Works

An agency’s stack is built almost entirely around the transaction. Capture the enquiry, win the listing, run the campaign, get it signed, get it settled. Every layer you pay for points at a property that is on the market right now.

Which is reasonable, because that is where the commission is. It also means the largest asset in the business gets no layer at all. The database holds every past vendor, every buyer who missed out, every appraisal that went to somebody else, and every enquiry that was not ready that month. It is the only part of the business that compounds, and it is maintained by whoever happens to remember.

The follow-up numbers are consistently grim wherever anyone measures them. Benchmark data from CRM vendors puts the average agent at around 1.4 follow-ups on a new lead before stopping, against seven to twelve contacts before a typical lead converts to an appointment, and fewer than 40 per cent of agents running any automated follow-up at all on leads older than ninety days. The most-quoted figure (that better than nine in ten past clients list their next property with somebody else) comes from US research, so treat the precise number with care. The mechanism is not country-specific.

2. The Layers of an Agency Stack

An agency stack follows the listing, not the org chart. Demand arrives, evidence is assembled, a pitch is made, a campaign runs, paper is signed, money moves, and then a relationship either continues or quietly ends. Every tool sits at one of those moments.

Seven of those eight layers have a mature Australian product category with several credible vendors. The eighth has a category name (database marketing) and mostly consists of newsletters nobody opens.

3. The CRM Is the Decision You Make Once

Changing CRM means migrating years of contact history, re-training every agent, rebuilding every portal feed and discovering which of your integrations were load-bearing. Agencies do it roughly as often as they rebrand, and for similar reasons.

The Australian field is genuinely competitive and segments cleanly by agency size. All of the majors handle local compliance and portal feeds, and all of them price by quote rather than a published seat rate, which is itself worth knowing before you start comparing.

4. What Agencies Actually Run

The names below are the ones that keep appearing in Australian agencies. As with any stack list, the last column matters more than the middle one: most agencies are over-tooled at the campaign end and under-tooled everywhere else.

Two notes. On portals, standard listings sit in the high hundreds and premium metro campaigns run into five figures, with the vendor generally carrying it as VPA whether or not the property sells. That makes portal spend the largest single line in most campaigns and the one vendors question hardest, which is a reason to be able to evidence its performance rather than assert it.

On AI content tools, adoption is close to universal: industry surveys put daily use across the great majority of agencies, overwhelmingly for listing copy, social posts and email drafts. It is worth being clear-eyed that this is the shallow end. Writing a listing description faster is a real saving and it changes nothing structural about the business.

5. Sales and Rentals Are Two Businesses

Most agencies run both and describe them as one. They share a brand, a shopfront, sometimes a CRM, and almost nothing else. The sales side is a campaign business: intense, episodic, commission-weighted, twelve weeks at a time. The rent roll is an operations business: continuous, compliance-heavy, and valued as an asset on its own.

The practical consequence for the stack is that a platform which is excellent at one is rarely excellent at the other, and the “one system for the whole agency” pitch usually means one of the two sides accepts a worse tool. That is sometimes the right trade. It should be a decision rather than a discovery.

6. Where the Value Leaks

Every tool in section 4 does its own job. The losses are between them, and in an agency they share a shape: a record exists, it is accurate, and no conversation ever happens off the back of it.

All six are the same failure wearing different clothes. The CRM knows everyone. An agent can hold maybe a few hundred live relationships in their head and their week. The difference between those two numbers is the whole opportunity, and it has never been a software problem because the missing step is a conversation, not a record. We have written up the mechanics of working a list like that (how to segment it, what to open with, and what a good outcome record looks like) in working the list you already own.

7. How to Choose

Six rules. None of them is about features, and the first one is worth more than the other five together.

1. 1

   Ask what happens on day 91

   For any tool touching leads or contacts, ask what it does with a record nobody has touched in three months. Most answers are "it sits there". That answer is the state of the industry and it is where your money is.

2. 2

   One contact record, defended

   The moment a second system holds a version of the contact (a marketing tool, a spreadsheet, an agent's phone), you have two truths and neither gets maintained. Everything reads from and writes to the CRM.

3. 3

   Buy for the appraisal, not the campaign

   Campaign tooling is mature, competitive and mostly interchangeable. The appraisal-to-authority step is where agencies actually differ, and it is chronically under-tooled relative to how much it decides.

4. 4

   Count what an agent does that is not talking

   Assembling vendor reports, chasing marketing approvals, re-keying between the data tool and the CRM. Every hour there is an hour not spent in the only activity that produces revenue.

5. 5

   Interrogate the word integration

   Which fields, in which direction, how often, and what happens when it fails. A portal feed that silently drops a photo set is a vendor phone call you will remember.

6. 6

   Every tool gets an owner

   In an agency this fails in a specific way: tools get bought by a principal and owned by nobody, so within a year each agent uses a personal subset and the reporting is fiction.

8. The Order to Build It In

Each layer inherits the discipline of the one below it. A proposal tool on top of a CRM nobody updates produces a beautiful pitch to a vendor whose last conversation was recorded wrong.

1. 1CRM, chosen for the size you will be in three years and then left alone
2. 2Portal feeds working reliably, because a broken feed is a vendor complaint with your name on it
3. 3Digital signing, which is the cheapest speed available to an agency
4. 4Property data, once you are defending prices rather than quoting them
5. 5Proposals, once you can state your appraisal-to-authority rate out loud
6. 6Reviews, systematically, from the tenth sale onward
7. 7Campaign and marketing production, once it has accidentally become somebody's job
8. 8Vendor reporting, automated from the sources rather than assembled by hand
9. 9Then the database: because everything above it feeds the database, and it only works once the records are worth acting on

The database sits last in build order and first in value, which is an uncomfortable combination and the reason most agencies never reach it. If you only ever do one thing out of order, do this one.

9. What Not to Buy Yet

10. Getting Started

1. 1

   Count the database honestly

   How many contacts, and how many have had a genuine two-way conversation in the last twelve months? The ratio is usually the most sobering number in the business and it takes an afternoon to produce.

2. 2

   Pull every appraisal you did not win

   Last three years, with the date and the reason if anyone recorded one. This list is the highest-intent group you own and it is almost always sitting in the CRM untouched.

3. 3

   Pull every underbidder

   Everyone who made an offer or registered and missed out. They are pre-qualified, they are still looking or they bought elsewhere, and either answer is worth knowing.

4. 4

   Price one conversation

   Take your average commission, your appraisal-to-authority rate and your listing-to-sale rate, and work backwards to what one genuine conversation with a past contact is worth. It is usually a number that makes the rest of this obvious.

5. 5

   Work one list properly

   Not the whole database. One list (the lost appraisals, say) worked all the way through with real conversations and recorded outcomes. What you learn about why they went elsewhere is worth as much as the listings you win back.

Platform positioning in this paper reflects vendor-reported figures and Australian industry software guides current to August 2026; agency CRMs price by quote, so none of the usual public benchmarks apply. Follow-up and reactivation statistics come from CRM vendor benchmark reporting and US industry research where noted, and the precise figures should be read as indicative rather than local.