Mortgage broker fact find process improvement

A fact find does three things at once. It gathers the information you need to research a case, it demonstrates that you understood the client's circumstances and objectives, and it shapes the conversation itself. Most fact finds are optimised for the first job, tolerable at the second and actively bad at the third.

Improving one is not about adding fields. It is usually about reordering, cutting, and deciding which answers you will take verbally and which you will insist on seeing evidenced.

Symptoms that your fact find needs work

• Advisers keep a private list of extra questions they always ask because the form does not.
• The same clarification email goes out after almost every appointment.
• Cases get to submission and then unravel over something that was asked about but recorded loosely.
• File checkers keep raising the same query about objectives or affordability.
• Clients go quiet during the meeting because they are being interrogated rather than advised.
• Nobody can complete one in under an hour and nobody knows which part is the slow bit.

If two or more of those are true, the structure is the problem, not the effort.

Ask in the order the client thinks, not the order the system stores

Most systems store data in tidy blocks: personal details, then employment, then income, then expenditure, then property, then objectives. Clients do not experience their situation in that order. They arrive with a property, a deadline and a worry.

Opening with the objective — what are you trying to do, by when, and what would make this go wrong for you — changes the tone of everything that follows. It also gives you the frame against which every later answer means something. An income structure is only interesting relative to what the client is trying to buy.

If your system insists on a fixed order, take the objectives conversation first anyway and enter it afterwards. The interface should not dictate the interview.

Branch early on the things that determine everything else

A small number of answers change the entire rest of the fact find. Employment structure is the big one. A salaried employee with two years at the same employer needs a fraction of the questioning that a director drawing dividends across two companies does.

Identify your branch points and put them near the front:

• How income is earned, and how many separate sources there are.
• Whether there is any adverse credit history, and roughly when.
• Whether the deposit is fully saved, partly gifted, or coming from a sale.
• Whether anyone will live in the property who is not on the mortgage.
• Whether there is an existing property being retained, sold or let.
• Whether the client has a hard deadline.

Once you know these six things you know which version of the fact find you are running. Asking every client every question because one in twenty needs it is the main reason fact finds take too long.

Separate what you are told from what you have seen

A recurring weakness in broker files is that stated figures and evidenced figures sit in the same fields with no distinction. Three months later nobody can tell whether the income figure came from a payslip or from the client's estimate on the phone.

Mark the difference. Some systems support it directly; where they do not, a note convention works. It costs nothing at the time and saves an argument later when a lender's figure disagrees with yours.

The same applies to expenditure. Clients underestimate their outgoings, consistently and without any intent to mislead. If you record a stated committed expenditure figure and then check it against statements, record both and note the reconciliation.

Make vulnerability a normal part of the conversation

Asking whether anything is going on that might affect how the client wants to be communicated with, or how they make decisions, is a normal question if you ask it normally. It becomes strange only when it is bolted on at the end as a compliance formality.

Good practice here is mostly about phrasing and placement. Ask it in the context of communication preferences rather than as a health enquiry. Record what the client actually said and what you agreed to do differently, rather than only ticking a category. Note it where the next person to touch the case will see it before they contact the client.

The specific record-keeping expectations around vulnerability and around demonstrating good outcomes are set by the FCA and, if you are an appointed representative, by your network on top of that. Check both rather than assuming a field in your CRM covers it.

Move the right work before the meeting

Pre-appointment data collection is worth doing, but only for the parts that are genuinely mechanical: names, dates of birth, address history, employer details, account details, document uploads. These are boring, error-prone when captured verbally, and perfectly suited to a form or portal.

What should not be pushed to a form is anything involving a judgement or a trade-off. Clients answer objectives questions badly in writing. They give the answer they think is expected, and you lose the follow-up question that would have found the real constraint.

A reasonable split is: the client fills in the facts, you conduct the conversation, and the meeting starts with you reflecting their form back to them and asking what it missed.

Test it on the file, not in the meeting

The honest test of a fact find is whether someone who was not present can read the resulting file and understand what the client wanted, what their circumstances were, what options were considered and why one was chosen. Pull three recent completed cases and try it, ideally with someone who was not the adviser.

You will find gaps in the same two places nearly every time: the reasoning behind the recommendation, and anything the client said that did not have a field waiting for it.

Small changes that usually pay off

A short list to work through

• Cut fields that have never influenced a recommendation. Count them first if you doubt it.
• Add a free-text prompt asking what the client said that does not fit anywhere else.
• Put affordability and expenditure questions after rapport is established, not at the start.
• Agree a house convention for how figures are sourced and stick to it across advisers.
• Have advisers swap files once a month and read each other's, briefly.
• Rewrite any question you find yourself explaining out loud every single time.

None of this needs new software. It needs someone to sit down with the current form, a red pen and three real cases.