Quick answer
In online business lending, software usually handles data collection, verification and first-pass scoring — categorising bank transactions, checking ID and flagging risks. People typically review anything outside standard criteria, larger amounts and flagged files. Many quick declines happen because an application falls outside a lender's automated rules, not because a person judged the business. Choosing the right lender first matters.
Key points
- Software speeds up collection, checking and scoring — it doesn't replace judgement on complex files.
- Automated declines often reflect a mismatch with one lender's rules.
- Applying to many lenders after automated declines can add credit enquiries.
- Context and explanations only help if a person gets to read them.
- Usually automated
- Data extraction, ID, scoring
- Usually human
- Exceptions, larger amounts, flags
- Quick declines
- Often policy mismatches
- Our enquiry
- Read by a person
One of the biggest worries about online lending is that a computer will say no before anyone understands your business. It’s a fair concern — and partly true. Software does a lot of the work in online lending. But it does different work from what most people imagine, and knowing where the line sits between automated and human decisions can change how you apply.
What does software actually do in an online application?
Most online lenders automate the parts of the process that are repetitive and data-heavy:
| Task | Typically automated? |
|---|---|
| Pulling in bank data and extracting transactions | Yes |
| Categorising income, expenses, other lenders and tax | Yes |
| Checking ID documents against issuing-agency records | Yes, with human review of mismatches |
| Checking ABN, company and director details | Mostly |
| Retrieving credit reports | Yes, once you consent |
| Applying hard policy rules (minimum trading time, turnover, industry) | Yes |
| Producing a score or risk grade | Often |
| Judging context, explanations and unusual situations | Usually a person |
| Approving larger or exceptional loans | Usually a person |
So software is usually the gatekeeper and the analyst’s assistant. It sorts, summarises and filters. The judgement call on anything that isn’t straightforward still tends to involve a person — a credit analyst or credit manager.
Why do online declines happen so fast?
A decline within minutes usually means the application hit a hard rule. Common triggers include:
- trading history shorter than the lender’s minimum
- turnover below the lender’s threshold
- an industry the lender doesn’t fund
- credit file events beyond its tolerance
- bank-data flags such as frequent dishonours or many other lenders
- a mismatch between application answers and verified data
The important point: that rule belongs to that lender. A different lender might fund the same business comfortably. An automated decline is information about fit, not a verdict on your business. Our explainer on bank statement analysis shows which data flags are most common.
Why does applying everywhere backfire?
After one quick decline, it’s tempting to try the next lender, and the next. But if each application includes a credit check, each one can be recorded on your file. Later lenders — including their automated systems — may read a cluster of recent enquiries as a sign that others have said no, making a decline more likely. It becomes a loop. See business versus personal credit files for how enquiries appear.
The better approach is to work out where you fit before you apply. That’s exactly the role of a specialist: knowing which lenders accept which situations, and putting your application in front of one that can say yes.
Where does human judgement make the difference?
People add value where data alone can mislead:
- Explaining anomalies — a one-off deposit that inflates turnover, or a planned shutdown that depresses it.
- Weighing history — an old, resolved default against strong current trading.
- Understanding structure — trusts, multiple entities, recent ownership changes.
- Assessing security and exit — property values, sale timelines, refinance plans.
- Matching need to product — realising a business asking for a loan really needs invoice finance or a line of credit.
Those are the situations where a well-prepared explanation, read by the right person, can change the outcome. It’s also why every eBusiness Loans enquiry is read by a lending specialist before anything is matched.
How can you get a fairer look at your application?
- Be accurate. Mismatches between your answers and verified data are a leading cause of automated declines.
- Disclose issues upfront. ATO debt, other lenders, past defaults. Surprises trigger rules; disclosures invite context.
- Include explanations. A short note on anything unusual helps a reviewer, if one sees it.
- Choose the lender carefully. A lender whose policy fits your profile is more likely to pass you through the automated stage to a person.
- Don’t rapid-fire applications. Space them out, or better, get advice first.
Will automation take over completely?
Data tools keep improving, and lenders keep automating more of the routine work. But business lending involves a huge variety of situations — different industries, structures, seasons and securities — and the cost of getting a decision wrong is significant. For the foreseeable future, the realistic model is software that does the collecting and sorting, with people deciding the cases that matter most. If you’d like to see what that looks like from submission to funding, read what happens after you submit.
What does an illustrative review look like?
A three-year-old landscaping business (illustrative) applies to an online lender. The software extracts twelve months of transactions and flags two things: a large deposit in January and four dishonour fees in March. The automated score lands in a borderline band, so the file goes to an analyst. She sees the business’s note explaining that the January deposit was the sale of an old ute and that March was a wet month with no work, after which trading recovered strongly. She excludes the vehicle sale from turnover, accepts the explanation for March given the clean months since, and approves a smaller amount than requested with a shorter term. Without the note, the same file might have been declined on the flags alone.
Ready to have a person look at your situation?
If you’ve been declined online before, or you’re worried a system won’t understand your business, start with a conversation instead of an algorithm. Send a 60-second enquiry — no credit check is involved, your details stay with one specialist rather than being blasted to a list of lenders, and a real person will read your situation before recommending where to apply. Please answer the form accurately and mention anything unusual; that context is exactly what helps us find the right fit first time.
Frequently asked questions
Why was I declined online in minutes?
Fast declines usually mean the application hit a hard rule in that lender's system — minimum trading time, turnover, industry, credit score or a flag in the bank data. It reflects that lender's policy, not necessarily whether any lender would help.
Can I ask for a person to review my application?
Sometimes. Some lenders allow a manual review or will accept further information. It's worth asking, but it's often more effective to choose a lender whose criteria fit before applying.
Does a human look at every eBusiness Loans enquiry?
Yes. A lending specialist reads every enquiry and calls you before anything is matched to a lender.
Will an automated decline affect my credit file?
If the lender ran a credit check as part of the application, that enquiry can be recorded, even if the decision was automated. That's why it pays to check fit first.