Preview of the new creaid.ai. Not live yet. Yellow = to confirm with you.

For mortgage brokers · 10 min read

How Brokers Use Loan Data to Pick the Right Lenders for a Deal

Short answer

Brokers use public loan records to see which lenders actually closed loans like the one in front of them: the same property type, the same county or state, the same loan size. Instead of calling everyone they know, they rank lenders by recent, similar closings, check each lender's balance sheet for appetite, and send the deal to a short list of five to ten that fit.

A commercial bank building
A commercial bank building. Example photo: Joe Mabel, CC BY-SA 3.0, via Wikimedia Commons.

Why do brokers start with data and not with a contact list?

Every broker has a contact list. The trouble is that a contact list shows who you know, not who is lending. Lenders change direction: a bank fills up on one property type, a credit union enters a new county, a loan officer moves and the appetite moves with them. A list built from memory goes stale quietly.

Loan data answers a narrower and more useful question: who has put money into this kind of deal, in this place, at this size? A lender's marketing says what it would like to do. Its closed loans say what its credit committee approved. For a broker, that difference shows up in three ways:

  • Fewer wasted submissions. A lender that has never closed a loan near the deal's size seldom starts with yours.
  • Faster answers. Lenders respond quickly to deals that look like the ones they already hold.
  • Better client conversations. You can tell a borrower why each lender is on the list, with numbers.

Data does not replace relationships. It tells you which relationships to use for this deal and where you need to build a new one. The relationship side is covered in how brokers build a lender network.

What public loan records exist, and what does each one show?

Three kinds of public records cover a large share of commercial real estate lending. None of them is complete on its own. Together they give a broker a working map.

RecordWhat it showsUseful forBlind spots
Federal mortgage data on apartment-building loansLoans by lender and county: amount, purchase or refinance, building size range, and often the rateApartment deals: who lends in a county, at what sizeCovers loans on housing, not retail, office or industrial; some lenders are not required to report
Bank and credit union regulatory reportsEach institution's balance sheet: how much it holds in each loan category and how that changes from report to reportJudging appetite and capacity by property categoryShows totals, not individual loans or terms
SBA loan recordsLoans by lender, program, amount, location and the borrower's type of businessOwner-occupied deals and business properties such as hotels, storage and medical officesOnly SBA-guaranteed loans

The first record is loan level, which means each row is one closed loan. That lets you count how many loans a lender made in a county and find its typical loan size. The second is institution level. It cannot tell you about one deal, but it shows whether a bank is growing or shrinking its book of apartment loans, owner-occupied loans, other commercial real estate loans or construction loans.

A plain-language walk through the same sources from the borrower's side is in how to use public loan data to find the right lender.

How do you turn loan records into a short list?

  1. Define the deal in data terms. Property type, state and county, loan amount, purchase or refinance, and for apartments the number of units.
  2. Pull every lender that closed that kind of loan in the area. Start with the county. Widen to neighboring counties or the state if the list is thin.
  3. Filter by loan size. Keep lenders whose closed loans bracket your amount. A lender whose middle loan is far above or below yours is a poor fit.
  4. Rank by count. A lender with twenty similar loans has a program. A lender with one may have done a favor for a customer.
  5. Check appetite. For banks and credit unions, look at whether the relevant loan category on the balance sheet has grown or shrunk over recent reports.
  6. Add what data cannot see. Private lenders, funds and lenders you know are active but that do not appear in these records.
  7. Cut to five to ten names and write one line on why each is there.

How many lenders to include depends on the deal. A clean, stabilized property may need only a handful. A harder deal needs a wider net. The reasoning is in how many lenders to approach.

A worked example: a $2,400,000 apartment refinance

Say a client needs a $2,400,000 refinance on a 24-unit apartment building. The numbers below are hypothetical. The loan records for that county show 40 lenders that closed at least one apartment loan in the latest year of data.

  1. Size filter. 25 of the 40 closed at least one loan between $1,000,000 and $5,000,000.
  2. Activity filter. 12 of those 25 closed five or more apartment loans in the county.
  3. Purpose filter. 8 of those 12 closed refinance loans, not only purchases.
  4. Result. The list went from 40 lenders to 8, a cut of 80%.

Now compare three lenders that survived the first filter: The middle loan is the median: half of the lender's loans were larger and half were smaller. It is a better guide than the average, because one very large loan can pull an average far from what the lender normally does.

LenderApartment loans closed in the countyMiddle loan sizeRead
Lender A30$2,100,000Strong fit: active, and the deal sits in the center of its range
Lender B6$14,000,000Weak fit: active, but its usual loan is almost six times larger
Lender C18$900,000Possible: active, but $2,400,000 may be near its upper limit

The time saved is easy to estimate. A first call or email exchange takes about 20 minutes per lender. Contacting all 40 would take 800 minutes, or more than 13 hours. Contacting 8 takes 160 minutes, under 3 hours, and those 8 are the ones most likely to quote. One more check. Suppose Lender A is a bank with $400,000,000 of capital and $1,300,000,000 of commercial real estate loans. That is 325% of capital. Bank regulators give extra scrutiny to banks whose commercial real estate loans pass roughly 300% of capital, so a bank above that mark may be slowing down even though last year's count looks strong. That is a reason to call and ask, not a reason to drop it.

What do bank and credit union reports add?

Closed-loan counts look backward. Regulatory reports help you guess what a lender will do next. A broker reads a few lines:

  • Growth in the category. A bank whose apartment loan balance has risen for several reports in a row is still buying. A falling balance suggests it is letting loans run off.
  • Concentration. Commercial real estate loans as a share of capital. High concentration often means tighter terms or a pause.
  • Construction exposure. Banks heavy in construction loans often limit new ones first.
  • Size of the institution. A bank's legal lending limit to one borrower is tied to its capital, so a small bank cannot make a large loan alone.
  • Loan quality. A rise in past-due and non-accrual loans usually leads to a more cautious credit committee.
  • Credit unions. Their reports show commercial (member business) loan balances, which reveal whether a credit union is a real commercial lender or an occasional one.

The differences between these institutions are covered in bank vs credit union for a commercial real estate loan.

What can loan data not tell you?

Good brokers know the limits of their tools. Public loan records have several.

  • They lag. Records are published well after loans close. A lender active in the data may have changed course since.
  • They show approvals, not declines. You see what closed, not what a lender turned down or why.
  • They leave out most terms. Recourse, prepayment, reserves, covenants and fees are not in the data.
  • They miss some lenders. Many private lenders, debt funds and life companies do not appear in these records.
  • They miss property types. Loan-level federal mortgage data covers apartments, not retail, office or industrial.
  • They do not show the person. Appetite often lives with a specific loan officer or team, and the data names only the institution.
  • Small counts mislead. One or two loans in a county are not a pattern.

The fix is simple: use data to choose who to call, then use the call to confirm. Ask each lender whether it is still making that loan, at what size, and what would stop it. The outreach itself is covered in how to contact commercial loan officers.

How to use the data in outreach and with clients

Data improves the pitch as well as the list. A message that says the deal resembles loans the lender closed in the same county, at a similar size, gets read. A message sent to everyone does not. Keep the request short and lead with the facts a lender screens on, as described in how to write a loan executive summary.

With clients, a data-backed list sets expectations. You can show why you are approaching eight lenders and not thirty, and what loan size and rate range similar closed loans fell in. Be careful with the wording: closed-loan rates describe past loans, not an offer. Track the answers you get, too. Your own record of who quoted, who passed and why becomes private data that no public source has. Tools for that are compared in commercial mortgage broker software.

Loan records are a research aid, so confirm current terms directly with each lender, and have the borrower review them with their own counsel.

Where CREaiD helps

CREaiD's lender search is built on public loan records covering more than 7,000 banks, credit unions, mortgage lenders, SBA lenders, private lenders and funds. You pick the state, the kind of loan and the loan amount and see the lenders that made that kind of loan there. For apartment loans you can narrow to a county, to purchase or refinance, and to the size of the building. More on how brokers use it is on the mortgage brokers page.

Brian Gramlich
About the author

Brian Gramlich is the CEO and founder of Terrydale Capital, a commercial real estate finance firm based in Dallas, Texas, and the founder of CREaiD.

Key takeaways

  • Closed loans show what a lender's credit committee approved, which is a better guide than marketing or memory.
  • Three public sources do most of the work: federal mortgage data on apartment loans, bank and credit union regulatory reports, and SBA loan records.
  • Filter by location, loan size, purpose and number of similar loans to cut a long list to five to ten lenders.
  • Balance sheet reports add a forward view: growth, concentration and capacity.
  • Data lags and leaves out terms and many private lenders, so every name on the list still needs a call.

Questions people ask

Where do commercial mortgage brokers find data on which lenders are active?

From public records and their own deal history. Federal mortgage data shows apartment-building loans by lender and county, bank and credit union regulatory reports show each institution's loan balances by category, and SBA records show guaranteed loans by lender and business type. Brokers combine these with what they learn from their own submissions.

Is public loan data available for retail, office and industrial loans?

Not at the loan level in the federal mortgage data, which covers housing. For those property types, brokers rely on bank and credit union balance sheet reports, SBA records for owner-occupied deals, county deed records and direct lender contact.

How many lenders should a broker send a deal to?

For most deals, five to ten well-chosen lenders is enough to produce competing quotes. Sending a deal to dozens of lenders that do not fit wastes time and can make the deal look shopped. Harder deals may need a second round with a wider list.

Can loan data tell me what rate a lender will quote?

No. Where rates are reported, they describe loans that already closed, under the market conditions of that time. They help you compare lenders with one another, but the quote on a new deal depends on current pricing and the deal itself.

How current is public loan data?

It always trails the market, because records are published some time after loans close or after a reporting period ends. Treat it as evidence of a lender's habits and confirm current appetite with a call before you promise a client anything.

Does using loan data replace a broker's lender relationships?

No. Data narrows the field and points to lenders you may not know, and relationships get the deal read, explained and approved. The strongest position is a short list drawn from data, worked through people you know or introduce yourself to.

Find lenders for your deal in minutes

CREaiD shows the lenders that really made your kind of loan in your state, from public loan records.

Start free