Reports / July 2026
CRE Lending Pulse — July 2026
In July 2026, RefiLoop indexed 3,945 commercial mortgage recordings across 60 counties in Texas, Ohio, Pennsylvania and Colorado. Every figure in this report is a count of recorded instruments — deeds of trust and mortgages classified as commercial — pulled directly from county records. We rank lenders by recording count, not dollar volume: the county indexes we cover did not state a loan amount on a single July instrument (est_amount is populated on 0 of 3,945 commercial recordings), and we do not estimate what the clerk does not state. All counts reflect newly recorded instruments — originations of record — not maturities, payoffs or loan balances.
One framing note before the tables: everything below describes activity among the recordings we index in these 60 counties. It is not a market total, and no figure here should be read as market share.
Coverage note: the first full-month edition
July is the first edition built on a full calendar month of full-footprint coverage (July 1–31). The June 2026 edition was a partial release — collection began June 25 and covered Harris County, Texas only, with 212 commercial recordings. That is why this report makes no month-over-month comparisons: June's 212 and July's 3,945 measure different footprints and different spans, not a change in activity. Meaningful trend lines start with the August edition.
The month in numbers
| Measure | July 2026 |
|---|---|
| Total recordings indexed (all classifications) | 9,958 |
| Commercial recordings | 3,945 |
| Unknown / residential recordings | 6,013 |
| Counties covered | 60 |
| States covered | 4 (TX, OH, PA, CO) |
| Distinct lenders matched | 676 |
| Commercial recordings matched to a tracked institution | 2,565 of 3,945 (65%) |
| Commercial recordings with a stated dollar amount | 0 |
| Period | 2026-07-01 to 2026-07-31 |
By state, the 3,945 commercial recordings break down as: Texas 2,543 · Ohio 1,026 · Pennsylvania 222 · Colorado 154. That split reflects our county footprint as much as anything else — Texas is where we cover the most counties.
Most active lenders — banks & depository institutions
Among recordings we index in these 60 counties, the most active banks and depository institutions of record in July were:
| # | Lender | Commercial recordings |
|---|---|---|
| 1 | Prosperity Bank (TX) | 56 |
| 2 | SSB Community Bank (OH) | 54 |
| 3 | Frost Bank (TX) | 47 |
| 4 | First Financial Bank (AL) | 39 |
| 5 | JPMorgan Chase Bank (OH) | 38 |
| 6 | Huntington National Bank | 35 |
| 7 | SouthState Bank NA | 27 |
| 8 | First Federal Community Bank, SSB (TX) | 25 |
| 9 | First Commonwealth Bank (PA) | 23 |
| 10 | City Bank (TX) | 20 |
Most active lenders — hard-money & investor lenders
Among the same indexed recordings, the most active non-bank lenders of record were:
| # | Lender | Commercial recordings |
|---|---|---|
| 1 | Kiavi Funding Inc | 80 |
| 2 | RFLF 4 LLC | 48 |
| 3 | Construction Loan Services II LLC | 22 |
| 4 | SCF Jake LP | 20 |
| 5 | Crebrid LLC | 18 |
| 6 | RBI Private Lending LLC | 15 |
| 7 | BPL Mortgage LLC | 15 |
| 8 | IFP Fund I LLC | 14 |
| 9 | Capital Fund I LLC | 14 |
| 10 | Schumer-Yarbrough Development LLC | 14 |
Why two tables? Because a combined ranking would mislead. Hard-money and investor lenders typically record many smaller, shorter-term instruments — bridge, fix-and-flip and construction-adjacent loans — so they generate far more recordings per dollar deployed than a bank writing term commercial mortgages. In a single combined table, Kiavi Funding (80 recordings) and RFLF 4 LLC (48) would occupy the top two spots and bury every depository institution, implying a comparison the data cannot support. Since we count recordings, not dollars, the honest presentation is to rank the two lending models separately.
Where the recordings are
The table below groups the top 20 counties by state. Read the counts as coverage, not as a heat map. How many commercial instruments we detect in a county depends heavily on the quality of that county clerk's index — how instruments are typed, how parties are named, how consistently documents are categorized — and that varies widely from clerk to clerk. Differences between counties in this table reflect detection and index quality at least as much as underlying lending activity, which is why we do not publish a "hottest counties" ranking.
| State | County | Commercial recordings |
|---|---|---|
| Texas | Harris | 1,098 |
| Texas | Montgomery | 257 |
| Texas | Denton | 208 |
| Texas | Collin | 119 |
| Texas | Brazos | 85 |
| Texas | McLennan | 75 |
| Texas | Hidalgo | 70 |
| Texas | Smith | 60 |
| Texas | Bell | 59 |
| Texas | Nueces | 59 |
| Texas | Grayson | 52 |
| Texas | Jefferson | 52 |
| Ohio | Franklin | 366 |
| Ohio | Cuyahoga | 224 |
| Ohio | Stark | 142 |
| Ohio | Butler | 109 |
| Pennsylvania | Berks | 121 |
| Pennsylvania | Delaware | 89 |
| Colorado | Arapahoe | 108 |
| Colorado | Boulder | 46 |
Property-type mix
Property-type classification runs after ingestion, and most July instruments have not yet been through it. We publish the raw counts anyway rather than report percentages of a small classified subset as if they described the whole month:
| Property type | Commercial recordings |
|---|---|
| Unspecified / not yet classified | 3,323 (84%) |
| Land | 492 (12%) |
| Industrial | 56 |
| Office | 38 |
| Multifamily | 21 |
| Retail | 15 |
The visible tilt toward land among classified instruments is consistent with what land deals look like in clerk indexes — they are often the easiest to classify from the recorded document alone — so treat the classified rows as an early, partial read, not a mix of the market.
Methodology & sources
What we count. Recorded commercial mortgage instruments — deeds of trust (TX, CO) and mortgages (OH, PA) classified as commercial by our pipeline — collected directly from county recorded documents in 60 counties. July 2026 figures cover instruments recorded July 1–31 and were re-verified against our production database on August 3, 2026.
Counts before dollars. No July instrument in our index states a loan amount (est_amount populated on 0 rows), so this edition contains no dollar figures. We attach dollar amounts only where the recorded instrument states them.
Originations of record, not maturities. A recording marks a new instrument hitting the county record. Nothing here reflects loan maturities, payoffs or outstanding balances.
Lender matching. 2,565 of 3,945 commercial recordings (65%) were matched to one of 676 tracked lending institutions. Unmatched recordings — individual lenders, single-purpose entities and institutions not yet in our registry — are included in county and state totals but absent from the lender tables.
Independence. All data comes from county recorded documents and federal bank Call Reports. No lender submissions, no paid placement, no opt-outs, no third-party aggregator data. Free to cite with attribution.
Suggested citation: RefiLoop Lender Data, CRE Lending Pulse — July 2026 (recorded documents, 60 counties in TX/OH/PA/CO). https://lenders.refiloop.com/reports/2026-07
Media contact: press@refiloop.com · https://lenders.refiloop.com
Source: Harris County recorded documents (deeds of trust) and federal bank Call Reports. Recording counts are county-record-verified; dollar amounts shown only where stated on or extracted from the instrument. Not an offer of credit or a recommendation of any institution.