Home → Insights

The Seven-Step Loan Application Nobody Should Have to Use

What follows is a composite. It is not one institution, and I have deliberately not named any. It is assembled from publicly accessible loan journeys across the region — the kind anyone can walk through from a phone, without credentials, without permission, without anybody knowing they are being observed.

I mention that because the individual steps below are not unusual. Every one of them exists somewhere. What makes the composite worth writing down is that they so often exist together, in the same journey, in institutions that would describe themselves as having gone digital.

The seven steps

One. You find the loan page. It explains the product well — rates, terms, eligibility, a photograph of a family outside a house. What it does not do is tell you how to apply. The button says Learn More.

Two. You find Apply. It opens an enquiry form: name, email, phone, and a free-text box asking what you are interested in. This is not an application. It is a request to be contacted about an application, though nothing on the page says so.

Three. If there is an online application, it requires an account first. So you register — a second set of credentials, a verification email, a password policy, a confirmation screen. None of this has anything to do with borrowing money.

Four. You begin the application, and you are asked for your date of birth, your address, your employer, and your account number. You have been a member for eleven years. The institution holds all of this. You type it in again.

Five. Identity verification arrives without warning. Upload a government-issued ID. There is no indication of accepted formats, no explanation of why it is needed at this point rather than at the start, and no way to save and return. Your session has a timeout you were not told about.

Six. Supporting documents. A recent payslip. A job letter. Two months of statements. None of this was mentioned on the product page, in the enquiry form, or at the start of the application. You are now seven or eight minutes in, and you do not have any of it to hand.

Seven. You submit what you can. The confirmation reads: a representative will contact you shortly. Some days later, you are asked to come into the branch to complete the process.

The seventh step was always a branch visit

That last step is the one worth sitting with, because it is not a failure of the journey. It is the design. The application was never capable of originating a loan on its own. It collects an expression of interest, formats it, and hands it to a member of staff, who then does the work that was always going to be done in person.

This matters more than the individual friction points above it. A slow form can be improved. A journey that ends in a branch visit by design cannot be improved — it can only be rebuilt, and rebuilding it is a decision somebody has to actually make.

Why the usual argument does not work here

The standard case for fixing this is abandonment: customers give up, customers leave, you lose them to a competitor. For a commercial bank in a contested market, that argument is sound.

For a member-owned credit union it is close to meaningless, and I think this is where a lot of well-intentioned digital advice goes wrong in our region.

A member is a shareholder. They have a share account, very likely a loan, often a family history with the institution going back a generation. When the online application defeats them, they do not leave. They drive to the branch. They join the queue. They are served, patiently and well, by someone whose time costs the institution money.

The member is not lost. The cost is entirely real — it simply lands on operations rather than on acquisition, which is why it rarely shows up in the conversations where digital investment gets decided. Nobody reports it. There is no line item called branch visits caused by a broken form.

Three numbers worth having

If abandonment is the wrong headline metric for a co-operative, these three are better, and none of them require new systems to start estimating.

Branch deflection failure. How many in-branch transactions exist only because a digital journey broke? Ask three tellers to keep a tally for a fortnight. Cost it in staff hours. This number is usually large and always uncomfortable.

Digital origination completion rate. Of the members who begin an application online, how many finish without human intervention? Not how many start. How many finish. In a journey like the one above, the honest answer is often zero, because step seven guarantees it.

Regulatory exposure in the journey. Where does KYC sit in the sequence? When are disclosures presented — before the member commits, or after? Is suitability assessed at a point where the answers can still change the outcome? This one has a named owner somewhere in your institution, because somebody answers to the regulator for it.

The question to ask this week

Before any of this becomes a project, there is a smaller and more revealing question, and it takes about a minute to ask:

Who here is accountable for how many members complete an application online?

If a name comes back immediately, you have someone to hand the three numbers to, and the rest is ordinary work. If the question produces a pause and a look around the table, that is the finding. A number nobody owns is a number nobody improves, and no amount of redesign survives contact with that.

It is worth knowing which of those two situations you are in before you spend anything.