The ideal mortgage broker software stack

Most brokerages assemble their technology by accident. Something is bought to solve an urgent problem, it stays, and three years later nobody can explain why the firm pays for two products that overlap. It helps to think of the stack as layers instead of a shopping list, because layers tell you what has to exist and what is optional.

There are five layers worth naming: acquisition, advice, submission, evidence, and oversight. Every brokerage has all five whether or not it has bought software for them. A solo adviser working from a notebook still acquires enquiries, gives advice, submits applications, keeps evidence and reviews performance. Software only decides how much of that is manual.

Layer one: acquisition

This layer takes a stranger and turns them into a booked appointment. It covers your website enquiry form, any directory or comparison listings you appear on, your telephone handling, and the calendar link clients use to book.

The failure mode here is a gap between the form and the diary. An enquiry arrives by email, sits in an inbox, and the adviser replies six hours later to find the client has already spoken to someone else. If you fix nothing else in this layer, connect the enquiry route directly to a bookable slot and to whatever system holds the client record.

Layer two: advice

Fact find, sourcing, criteria checking, affordability, protection quotes, and the suitability report all sit here. This is the layer with the most vendors and the most confusion, because some products cover all of it and some cover one slice.

The question that matters is how many times a client's income figure gets typed. In a badly assembled stack it is typed into the fact find, again into the sourcing tool, again into the lender portal, and again into the suitability letter. Each retype is a chance to introduce an error you will have to explain later.

Layer three: submission

Lender portals, packaging, and the mechanics of getting a case to a decision. This layer is largely outside your control. Lenders own their portals and they do not integrate uniformly. Some brokerage platforms offer direct submission to a subset of lenders; others simply hold your case notes while you work in the lender's own system.

Be realistic about what any vendor can promise here. Ask which specific lenders are covered by any submission integration, and ask when that list was last updated, because coverage changes.

Layer four: evidence

Documents, identity and anti-money-laundering checks, consent records, call recordings, disclosure documents and the audit trail behind each recommendation. This layer is not glamorous and it is the one that decides whether a file review goes well.

Under Consumer Duty a firm needs to be able to show not just what was recommended but that it considered the client's needs and that the client understood the outcome. That is an evidence problem before it is a technology problem, but the technology determines how long it takes to produce the evidence when someone asks.

Layer five: oversight

Pipeline visibility, adviser performance, income reconciliation against procuration fees, complaints logging, and the management information a principal needs. In a two-person firm this can honestly be a monthly conversation and a spreadsheet. Past roughly five advisers it stops being possible to hold in your head.

What order to buy in

If you are building from close to nothing, the sequence that tends to cause least pain is: client record first, documents second, diary and reminders third, reporting last.

Client record first, because everything else attaches to it. If you buy a document tool before you have a place to put clients, you end up with folders organised by surname and no way to answer "which cases are with Halifax right now".

Reporting last, because reporting is only as good as the data underneath it. Firms that buy a dashboard early get a very attractive picture of incomplete data.

Where the money usually goes wrong

Overlap is the common one. A firm pays for an all-in-one platform and also keeps a general-purpose CRM, because two advisers preferred the old one. Both get half-used and neither is trustworthy.

The second is per-seat pricing on tools that only one person touches. Reporting and compliance tools often only need principal access, but firms buy firm-wide licences by default.

The third is buying for the firm you plan to be. A platform sized for twenty advisers, bought by a firm of three that intends to grow, usually costs two years of subscription before the growth arrives, and by then the market has moved anyway.

Constraints that are not yours to choose

If you are an appointed representative, your network may mandate the CRM, the sourcing system, the suitability template, or all three. That is not always a bad deal, since the network absorbs the cost and the compliance configuration, but it does mean the stack conversation starts with what the network supplies and what it permits alongside.

Directly authorised firms have more freedom and correspondingly more responsibility. You choose the tools, you configure the compliance workflow, and you own the consequences if the audit trail has holes.

Either way, check data portability before you sign anything. Ask what format an export comes in, whether it includes documents and notes or only contact fields, and how long you have to retrieve data after cancelling.

A short honest note on vendors

Named platforms in this market — Acre, Smartr365, Twenty7Tec, IRESS and others — each publish their own feature lists, and those lists change. Anything written about specific capability or price ages quickly. Take the layer model above into a demo, ask the vendor to show your own workflow rather than their standard script, and confirm current pricing and integrations directly with them.

The stack that works is rarely the one with the most features. It is the one where an adviser can answer, in under thirty seconds, what the next action is on any live case.