Real estate deal pipeline screening: 8 of 31 deals pass the gates

An acquisitions analyst hands Compound a pipeline to triage. It pulls 31 Midwest properties whose CMBS loans mature by November 2028, prices each at its latest appraised value, and puts all of them through one set of loan terms, three gates and a shortlist that fills by formula.

Compound Research · Updated Nov 15, 2026 · 14 min read

Short answer

Real estate deal pipeline screening puts every deal on one row under shared assumptions, then gates and ranks them. In Compound's screen of 31 Midwest properties with CMBS loans maturing by November 2028, each priced at its appraised value, 8 pass all three gates and 23 show negative leverage against a 7.90% loan constant.

Try it now
On this page (9)
  1. 1. The shared assumptions: one loan for every deal
  2. 2. The pipeline: 31 properties from CMBS filings
  3. 3. Growth by property type: sourced for apartments, assumed for the rest
  4. 4. Going-in cap rate against the 7.90% loan constant
  5. 5. The gates and the rank by levered IRR
  6. 6. Refinance shortfall and appraisal age flags
  7. 7. What the public data cannot tell you
  8. 8. Our view: use the shortlist as a reading order
  9. 9. FAQ

Price 31 Midwest properties at their latest appraised value, finance each with a new 6.90% loan, and 23 of them earn less on the purchase price than the loan costs each year. Eight pass all three gates of the screen. One of those eight returns less than the equity put in.

You are on the acquisitions team of an institutional investor and someone has handed you a pipeline. Before anyone builds a full model, you need every deal on one page, on the same loan terms, with the same tests. We asked Compound to build that page from public filings. It ran on November 15, 2026 and returned one Excel workbook with three sheets.

What the workbook shows:

  • 8 of 31 deals pass. The three gates are a year-1 debt service coverage ratio of at least 1.25x, a going-in cap rate at or above the cost of the new loan, and occupancy of at least 80%.
  • One gate decides every result. All 23 deals that fail have a going-in cap rate below the 7.90% annual loan constant. The coverage and occupancy gates remove no deal that the leverage gate keeps (our count from the gate columns).
  • The top of the shortlist is a warehouse and five apartment properties. Bolingbrook Logistics Center ranks first with a 16.8% levered IRR in the model. No retail property passes.
  • A pass is a filter result. The gates do not test return. One office deal passes all three and ranks eighth with a modeled levered IRR of -7.9% and an equity multiple of 0.73x.
  • 30 of 31 appraisals are more than three years old. The asking price in this screen is the appraised value, so the age of the appraisal is the age of the price.
  • Five deals carry a refinance shortfall flag. In each, a new loan at 65.0% of appraised value is smaller than the existing loan balance.

Every return on this page is the output of a screen at an assumed price. The request set each asking price at the latest appraised value in the loan filing. No seller has quoted these prices, and nothing here is a forecast for any property or a view on whether to buy it.

The property data comes from CMBS loan-level filings, the Form ABS-EE exhibits that each trust files on SEC EDGAR. The loan rate starts from the Daily Treasury Par Yield Curve. Expense growth is the consumer price index from the Bureau of Labor Statistics. Apartment rent growth blends Zillow’s ZORI index with Apartment List rent reports that Compound read on the web. The Sources sheet lists 25 trusts, each with its filing accession number.

Prompt to try

“I have been handed a pipeline of deals to triage and want a deal pipeline screening workbook that ranks them in one sitting, built on real properties. From the CMBS loan-level filings, pick 25 to 40 properties in the Midwest (Ohio, Michigan, Indiana, Illinois, Wisconsin, Minnesota and Missouri) whose loans mature within the next 24 months, since those owners must sell or refinance. Mix apartment, industrial, retail and office, and keep only properties whose latest report gives units or net rentable area, occupancy, net operating income, appraised value and the loan balance, rate and maturity date. Give each deal one row with its name, city and state, property type, year built, size, occupancy, reported NOI and an asking price set at the latest appraised value, plus one capital figure per unit or per square foot as an analyst assumption by property type and an optional price override. Set the assumptions once for every deal in a block at the top: loan-to-value, a loan rate built from today's 10-year Treasury yield plus a spread, amortization, exit cap rate spread over the going-in rate, hold period, rent growth by property type (apartments from Zillow ZORI and Apartment List for the Midwest metros, the others as analyst assumptions) and expense growth from the BLS CPI. Build the screen itself as a one-sheet model: the shared assumptions block on top, then one table with one row per deal and the computed columns to the right of the inputs: price per unit or per square foot, going-in cap rate, year-1 DSCR on the new loan, year-1 cash-on-cash, levered IRR, equity multiple, and whether the new loan at the shared terms pays off the existing balance at maturity. Add pass or fail gates (a DSCR of at least 1.25x, no negative leverage, occupancy of at least 80%), a rank by levered IRR among the deals that pass, and a top-ten shortlist block that fills by formula and re-sorts when I change a shared assumption. Flag every deal with negative leverage, every deal where the new loan falls short of the existing balance, and every deal whose appraisal is more than three years old. Use the stored public data wherever it has the figure, and web search only for what it lacks. Put every input on an Assumptions sheet with one column that says where it came from: a cited source, or "analyst assumption" for a figure I would set myself, with the reason for the value you chose. Every output must be a live formula that traces back to the Assumptions sheet. Add a Sources sheet listing each data source, what it gave and its date. Deliver one Excel workbook and a short summary of the result, the flags, and what I would need from the seller to firm it up. Cite every sourced figure.”

Try this prompt free →

That is the request as we sent it, without its closing line, which asks that no name or email be put in the workbook.

The shared assumptions: one loan for every deal

Every deal in the screen gets the same loan: 65.0% of the price at 6.90%, amortizing over 30 years. Those terms produce an annual loan constant of 7.90%, and that one number drives most of the results.

Compound put the shared terms in one block on the Assumptions sheet and pointed the top of the Screen sheet at it. Each row says where its value came from.

Midwest_CMBS_Deal_Pipeline_Screen.xlsx
C15fx=-PMT(C12/12,C14*12,1)*12
BCDE
6AssumptionValueUnitSource / basis
7Analysis date2026-11-15dateRequest: the screen date; the maturity window and appraisal age are measured from it.
8Loan maturity window24monthsRequest: loans maturing within 24 months of the analysis date.
910-year Treasury par yield, close of Nov 13, 20265.15%, as publishedU.S. Treasury daily par yield curve, 10-year, as of 2026-11-13, the latest close before the analysis date (Compound stored rates data).
1010-year Treasury yield5.15%%Formula: published yield / 100.
11Loan spread over 10-year Treasury175bpsAnalyst assumption: 175 bps is a typical fixed-rate CMBS / balance-sheet spread for stabilized Midwest assets at 60-65% LTV.
12New loan interest rate6.90%%Formula: 10-year Treasury + spread.
13Loan-to-value65.0%%Analyst assumption: 65% is a common senior-loan ceiling for stabilized multifamily, industrial, retail and office.
14Amortization30yearsAnalyst assumption: 30-year amortization, standard for fixed-rate commercial mortgages.
15Annual loan constant7.90%%Formula: annual debt service per $1 of loan at the new rate and amortization.
16Closing costs2.0%% of priceAnalyst assumption: 2% for transfer taxes, title, legal and financing fees.
17Hold period5years (1 to 10)Analyst assumption: 5-year hold, a typical value-add / core-plus horizon.
18Exit cap rate spread over going-in cap50bpsAnalyst assumption: 50 bps of cap rate expansion over the hold for aging and rate risk.
19Sale costs2.0%% of exit valueAnalyst assumption: 2% for brokerage and closing costs at sale.
20Minimum year-1 DSCR1.25xxRequest: DSCR of at least 1.25x.
21Minimum occupancy80%%Request: occupancy of at least 80%.
22Appraisal age flag threshold3yearsRequest: flag appraisals more than three years old.
23Expense growth (CPI-U, 12 months to Sep 2026)3.7%%Formula: BLS CPI-U 12-month change from the CPI block below.
ScreenAssumptionsSources
The shared assumptions block from Compound's Assumptions sheet, as written. The Source / basis column separates what the request set, what a public source gave and what is an analyst assumption.

Three kinds of input sit in that block, and the label tells you which is which.

KindInputsWhat to do with it
RequestAnalysis date, 24-month maturity window, 1.25x minimum coverage, 80% minimum occupancy, three-year appraisal age limitThese came from the request. Change them when your mandate differs.
Sourced10-year Treasury yield of 5.15% at the close of November 13, 2026; expense growth of 3.7%Refresh on the day you rerun the screen.
Analyst assumption175 basis point loan spread, 65.0% loan-to-value, 30-year amortization, 2.0% closing costs, 5-year hold, 50 basis point exit cap spread, 2.0% sale costsReplace with your lender’s quote and your house view.

The loan rate is the Treasury yield plus the spread: 5.15% plus 175 basis points is 6.90%. The loan constant in cell C15 is the yearly payment on one dollar of that loan, 7.90%. It is higher than the rate because the payment includes principal.

The constant matters more than the rate in this screen. Compound’s test for negative leverage compares each going-in cap rate with the 7.90% constant. A deal bought at a cap rate below the constant pays out more in debt service per dollar borrowed than the property earns per dollar of price, so borrowing lowers the first-year cash return. Some analysts run that test against the interest rate, as the screen on the real estate deal screener page does. The constant is the stricter version.

The pipeline: 31 properties from CMBS filings

Compound kept 31 properties in six states: 9 apartment, 5 industrial, 6 office and 11 retail. Each one’s loan matures between February 2027 and November 2028, and each one’s latest filing reports size, occupancy, net operating income, appraised value and the loan’s balance, rate and maturity.

The request asked for 25 to 40 properties across seven states. Compound said in its answer that no Minnesota loan met every data requirement, and that it kept single-property loans that were current on payments and had 12 months of financials. The inputs come from each trust’s October 2026 filing.

Midwest_CMBS_Deal_Pipeline_Screen.xlsx
S26fx=IF(N(Q26)>0,Q26,M26)
BCDEFGHIJLMNOP
25IDPropertyCityStateTypeYear builtSizeUnitOccupancyReported NOIAppraised value (asking)Appraisal dateExisting loan balanceLoan maturity
26D01Centennial HighlandsFort WayneINApartment201597Units96.9%1,483,18218,550,000Jul 201711,371,761Oct 2027
27D02Chatham GlenPlainfieldINApartment201692Units95.7%1,211,38917,200,000May 201811,428,285Jul 2028
28D03Colony Bay ApartmentsFort WayneINApartment1966495Units96.0%2,573,90225,000,000Mar 201713,500,000Apr 2027
29D04Clearpointe LandingMidlandMIApartment2016110Units95.5%1,793,63020,900,000Jul 201813,400,000Nov 2028
30D05KINGSLEY APARTMENTSSterling HeightsMIApartment1969328Units95.0%2,975,67330,700,000May 201820,217,010Jul 2028
31D06River Place ApartmentsDetroitMIApartment1891301Units88.4%1,464,88232,800,000May 201817,000,000Jul 2028
32D07WOODLAND GARDENS APARTMENTSRoyal OakMIApartment1967337Units97.0%2,931,69134,100,000May 201822,677,633Jul 2028
33D08DRAKE PLAZA APARTMENTSSt. LouisMOApartment191478Units84.0%380,6615,900,000Jan 20183,197,881May 2028
34D09118 FLATS OVALClevelandOHApartment201632Units75.0%400,5588,100,000Jan 20174,904,020Mar 2027
35D10Bolingbrook Logistics CenterBolingbrookILIndustrial200742,350SF100.0%1,328,86314,100,000Apr 20175,772,802Jun 2027
36D11GRM IndianapolisIndianapolisINIndustrial1940394,570SF100.0%2,165,66930,100,000Mar 202317,000,000May 2028
37D12Novi Technology CenterNoviMIIndustrial1988105,184SF84.7%907,30312,500,000Jan 20177,805,340Apr 2027
38D13MAPLE STREETNorth CantonOHIndustrial1919788,032SF79.0%2,505,05334,100,000Mar 201820,759,438Jul 2028
39D14NORTHERN OHIO INDUSTRIAL PARKElyriaOHIndustrial19461,056,528SF88.0%2,148,67015,900,000Jun 20179,165,518Nov 2027
40D15Bingham Office CenterBingham FarmsMIOffice1974524,852SF72.4%3,176,22752,000,000Mar 201726,012,637May 2027
41D16CONTINENTAL PLAZATroyMIOffice198695,953SF65.0%261,7649,200,000Nov 20165,264,765Mar 2027
42D17LIBERTY CENTERTroyMIOffice1988278,443SF97.0%3,666,65233,000,000Jan 201718,495,648Feb 2027
43D18RAYMOND JAMES FINANCIAL CENTERAuburn HillsMIOffice1987137,960SF83.0%1,047,54615,000,000Apr 20179,304,132Jun 2027
44D19Alliance Data SystemsColumbusOHOffice2017241,493SF100.0%3,111,62247,000,000Jul 201826,550,000Aug 2028
45D20Belcan HeadquartersBlue AshOHOffice2000135,413SF100.0%1,814,53726,500,000Sep 202115,900,000Nov 2028
46D21901 West MadisonChicagoILRetail200716,052SF77.1%449,4087,400,000Sep 20254,924,424Aug 2028
47D22DEMPSTER PLAZANilesILRetail1968113,191SF91.0%1,906,60336,150,000Aug 201722,366,703Oct 2027
48D23JEWEL OSCO FOX RIVERFox River GroveILRetail199055,553SF100.0%506,8529,000,000Jul 20185,100,000Oct 2028
49D24JEWEL-OSCOShorewoodILRetail200266,437SF100.0%752,38213,200,000Jul 20188,600,000Sep 2028
50D25WEST CARMEL MARKETPLACECarmelINRetail2005121,000SF100.0%1,508,62121,400,000Sep 20179,803,138Jan 2028
51D2621300 EurekaTaylorMIRetail201723,200SF100.0%571,8457,700,000Jul 20184,553,818Oct 2028
52D27Romeo CommonsMacombMIRetail198676,636SF98.0%1,060,37014,300,000Jan 20177,861,165Mar 2027
53D28Gravois Dillon PlazaHigh RidgeMORetail1996148,110SF96.0%1,098,49420,100,000Nov 201712,901,822Jan 2028
54D29MARKETPLACE AT FOUR CORNERS PHASE IIAuroraOHRetail2004116,436SF84.0%700,86816,310,000Jan 20189,819,147Mar 2028
55D30EAST TOWNE GREEN BAYGreen BayWIRetail198367,471SF96.0%732,23710,420,000Feb 20186,068,326Jun 2028
56D31PICK N SAVE APPLETON WIAppletonWIRetail200465,313SF100.0%738,15111,500,000Oct 20176,809,505Jan 2028
ScreenAssumptionsSources
The deal inputs on Compound's Screen sheet, all 31 rows, with 14 of the input columns. Values are as written; revenue, the price override and capital per unit are left out here. Names, sizes, occupancy, income, appraisals and loan terms are as each trust's Form ABS-EE filing reports them.

The formula in the bar is the purchase price. It takes the price override in column Q if you have typed one, and the appraised value otherwise. That is the one place where you turn a public appraisal into a seller’s actual ask.

Three things in this table are worth a look before any return. First, the size of the pipeline: the appraised values add up to $650.1 million and the existing loan balances to $378.5 million (our arithmetic on columns M and O).

Second, the appraisal dates. 27 of the 31 appraisals are dated 2017 or 2018 (our count). The Assumptions sheet says, deal by deal, whether the value is the most recent reported appraisal or the appraisal at securitization with no later one reported.

Third, the existing loans are cheap by today’s terms. The filings report note rates from 3.87% to 6.26% on the 31 loans, and the screen’s new loan costs 6.90%. Every one of these loans would be replaced by a more expensive one.

One mapping is Compound’s and it is disclosed: the filings code three of the industrial properties as warehouse, and the Assumptions sheet says it mapped that code to Industrial.

Growth by property type: sourced for apartments, assumed for the rest

Apartment rent growth is 3.1%, the average of two public rent measures across five Midwest metros. Industrial, office and retail growth are analyst assumptions of 3.0%, 0.0% and 2.0%, and expenses grow 3.7% for every type.

The request asked for apartment rent growth from Zillow ZORI and Apartment List, and for the other types as analyst assumptions. Compound built the apartment figure in a block you can audit.

Midwest_CMBS_Deal_Pipeline_Screen.xlsx
C41fx=AVERAGE(E39,F39)
BCDEF
33MetroZORI multifamily, Sep 2025 ($/mo)ZORI multifamily, Sep 2026 ($/mo)ZORI 12-month changeApartment List 12-month change
34Cleveland, OH Metro Area1,2011,2494.0%(0.7%)
35Detroit-Warren-Dearborn, MI Metro Area1,3461,3953.7%3.8%
36Fort Wayne, IN Metro Area1,0551,1337.3%5.0%
37Indianapolis-Carmel-Greenwood, IN Metro Area1,3531,3872.5%1.5%
38St. Louis, MO-IL Metro Area1,2831,3293.5%0.5%
39Average of the five metros4.2%2.0%
41Apartment rent growth (blend)3.1%
44SeriesValueUnit
45CPI-U all items, not seasonally adjusted, Sep 2025324.800index 1982-84=100
46CPI-U all items, not seasonally adjusted, Sep 2026336.940index 1982-84=100
47CPI-U 12-month change3.7%%
ScreenAssumptionsSources
The apartment rent growth block and the consumer price index block from Compound's Assumptions sheet, without the source column. Column headings are shortened; the values are as written.

The two measures disagree. ZORI averages 4.2% across the five metros and Apartment List averages 2.0%. In Fort Wayne they read 7.3% and 5.0%. In Cleveland ZORI shows 4.0% growth and Apartment List shows a 0.7% decline. Compound averaged the two averages and used 3.1% for every apartment deal. The ZORI figures come from stored data. The Apartment List figures came from its November 2026 rent report on the web, and the source note says four are city-level figures and Indianapolis is metro-wide.

Expense growth is sourced: the consumer price index rose from 324.800 in September 2025 to 336.940 in September 2026, a 3.7% change.

The other inputs by type are labeled analyst assumption, each with a reason.

TypeRent growthCapital at purchaseBasis
Apartment3.1%$5,000 per unitSourced blend; capital is an analyst assumption
Industrial3.0%$5 per SFAnalyst assumption
Office0.0%$25 per SFAnalyst assumption
Retail2.0%$10 per SFAnalyst assumption

Look at what those pairs do. Year-1 net operating income in the screen is revenue grown at the rent rate less expenses grown at 3.7%. For office, revenue is flat and expenses rise, so income falls every year of the hold. The office results below follow from that choice as much as from the filings.

Going-in cap rate against the 7.90% loan constant

23 of the 31 deals have a going-in cap rate below the 7.90% loan constant, and the screen flags each one for negative leverage. All 11 retail deals are flagged, with cap rates from 4.3% to 7.4%.

The going-in cap rate is reported net operating income divided by the purchase price. With the price set at appraised value, it is income as a share of an appraisal, and most of these appraisals are eight or nine years old.

Going-in cap rate at appraised value, 31 deals

Reported NOI divided by appraised value. Dark bar: the 7.90% annual loan constant. Highlighted bars pass all three gates

Northern Ohio Industrial Park, OH13.5%Liberty Center, MI11.1%Colony Bay Apartments, IN10.3%Kingsley Apartments, MI9.7%Bolingbrook Logistics Center, IL9.4%Woodland Gardens Apartments, MI8.6%Clearpointe Landing, MI8.6%Centennial Highlands, IN8.0%Annual loan constant on the new loan7.9%21300 Eureka, MI7.4%Romeo Commons, MI7.4%Maple Street, OH7.3%Novi Technology Center, MI7.3%GRM Indianapolis, IN7.2%West Carmel Marketplace, IN7.0%Chatham Glen, IN7.0%East Towne Green Bay, WI7.0%Raymond James Financial Center, MI7.0%Belcan Headquarters, OH6.9%Alliance Data Systems, OH6.6%Drake Plaza Apartments, MO6.5%Pick N Save Appleton WI, WI6.4%Bingham Office Center, MI6.1%901 West Madison, IL6.1%Jewel-Osco, IL5.7%Jewel Osco Fox River, IL5.6%Gravois Dillon Plaza, MO5.5%Dempster Plaza, IL5.3%118 Flats Oval, OH5.0%River Place Apartments, MI4.5%Marketplace at Four Corners Phase II, OH4.3%Continental Plaza, MI2.9%

Compound's Screen sheet, column U and cell D13. NOI and appraised values: CMBS Form ABS-EE filings for the reporting period ending October 2026. Loan constant: 10-year Treasury of 5.15% on November 13, 2026 plus an assumed 175 basis point spread, 30-year amortization.

The eight deals above the dark bar are the eight that pass.

A coverage ratio of 1.25x on a 65.0% loan with a 7.90% constant needs a going-in cap rate of about 6.42% (our arithmetic: 1.25 times 65.0% times 7.90%, before year-1 growth). The leverage gate needs 7.90%. Any deal that clears the leverage gate has already cleared coverage by a wide margin. So the coverage gate fails 10 deals and the occupancy gate fails five, and every one of them also fails the leverage gate (our counts from columns AK, AL and AM). Of the 23 deals that do not pass, thirteen cover their debt at 1.25x or better, and twelve of those miss on leverage alone.

By property type (ranges from column U, pass counts are ours):

TypeDealsGoing-in cap rate rangePass all gates
Apartment94.5% to 10.3%5
Industrial57.2% to 13.5%2
Office62.8% to 11.1%1
Retail114.3% to 7.4%0

Now the question of how strict the test is. 14 of the 23 flagged deals have a cap rate below the 6.90% interest rate, and nine sit between the rate and the constant (our counts). If your shop defines negative leverage against the rate, those nine deals come off the flag list. Compound’s answer makes a related point: cutting the loan spread to 100 basis points lifts the pass count to 10. The gate is one formula in column AL that points at one cell, so either change takes a few seconds.

Read every row of the next table as a hypothetical purchase. Each one assumes a new 65.0% loan at 6.90% and a price equal to an appraisal that is, for 30 of the 31, more than three years old, with one growth rate and one capital figure for the whole property type. A FAIL in the Result column or a negative return describes that set of assumptions. It does not describe how the property or its existing loan is performing.

Midwest_CMBS_Deal_Pipeline_Screen.xlsx
AL26fx=IF(U26>=$D$13,"PASS","FAIL")
CFTUAAABAFAGAHANAOAPAQAR
25PropertyTypePrice / unit or SFGoing-in cap rateYear-1 DSCRYear-1 cash-on-cashLevered IRREquity multipleNew loan less existing balanceResultRank (levered IRR)Negative leverageRefi shortfallStale appraisal
26Centennial HighlandsApartment191,2378.0%1.60x7.8%12.0%1.65x685,739PASS5STALE APPRAISAL
27Chatham GlenApartment186,9577.0%1.41x5.3%8.7%1.46x(248,285)FAILNEG LEVERAGEREFI SHORTFALLSTALE APPRAISAL
28Colony Bay ApartmentsApartment50,50510.3%2.06x11.6%12.8%1.65x2,750,000PASS4STALE APPRAISAL
29Clearpointe LandingApartment190,0008.6%1.72x9.3%13.5%1.73x185,000PASS3STALE APPRAISAL
30KINGSLEY APARTMENTSApartment93,5989.7%1.94x11.4%14.5%1.77x(262,010)PASS2REFI SHORTFALLSTALE APPRAISAL
31River Place ApartmentsApartment108,9704.5%0.88x(1.5%)(4.8%)0.78x4,320,000FAILNEG LEVERAGESTALE APPRAISAL
32WOODLAND GARDENS APARTMENTSApartment101,1878.6%1.72x8.8%11.8%1.62x(512,633)PASS6REFI SHORTFALLSTALE APPRAISAL
33DRAKE PLAZA APARTMENTSApartment75,6416.5%1.28x3.3%3.2%1.16x637,119FAILNEG LEVERAGESTALE APPRAISAL
34118 FLATS OVALApartment253,1254.9%0.99x(0.1%)2.8%1.15x360,980FAILNEG LEVERAGESTALE APPRAISAL
35Bolingbrook Logistics CenterIndustrial3339.4%1.89x11.8%16.8%1.93x3,392,198PASS1STALE APPRAISAL
36GRM IndianapolisIndustrial767.2%1.44x5.2%7.7%1.40x2,565,000FAILNEG LEVERAGESTALE APPRAISAL
37Novi Technology CenterIndustrial1197.3%1.45x5.6%8.2%1.43x319,660FAILNEG LEVERAGESTALE APPRAISAL
38MAPLE STREETIndustrial437.3%1.46x4.9%3.3%1.16x1,405,562FAILNEG LEVERAGESTALE APPRAISAL
39NORTHERN OHIO INDUSTRIAL PARKIndustrial1513.5%2.69x12.4%7.8%1.35x1,169,482PASS7STALE APPRAISAL
40Bingham Office CenterOffice996.1%1.14x1.1%(41.9%)0.08x7,787,363FAILNEG LEVERAGESTALE APPRAISAL
41CONTINENTAL PLAZAOffice962.8%0.48x(4.2%)n/a-1.07x715,235FAILNEG LEVERAGESTALE APPRAISAL
42LIBERTY CENTEROffice11911.1%2.11x9.8%(7.9%)0.73x2,954,352PASS8STALE APPRAISAL
43RAYMOND JAMES FINANCIAL CENTEROffice1097.0%1.30x2.6%(33.3%)0.18x445,868FAILNEG LEVERAGESTALE APPRAISAL
44Alliance Data SystemsOffice1956.6%1.28x2.9%(7.3%)0.70x4,000,000FAILNEG LEVERAGESTALE APPRAISAL
45Belcan HeadquartersOffice1966.8%1.29x3.0%(21.0%)0.35x1,325,000FAILNEG LEVERAGESTALE APPRAISAL
46901 West MadisonRetail4616.1%1.20x2.6%0.9%1.04x(114,424)FAILNEG LEVERAGEREFI SHORTFALL
47DEMPSTER PLAZARetail3195.3%1.04x0.4%(3.5%)0.84x1,130,798FAILNEG LEVERAGESTALE APPRAISAL
48JEWEL OSCO FOX RIVERRetail1625.6%1.12x1.4%0.4%1.02x750,000FAILNEG LEVERAGESTALE APPRAISAL
49JEWEL-OSCORetail1995.7%1.13x1.6%1.2%1.06x(20,000)FAILNEG LEVERAGEREFI SHORTFALLSTALE APPRAISAL
50WEST CARMEL MARKETPLACERetail1777.0%1.39x4.7%2.6%1.12x4,106,862FAILNEG LEVERAGESTALE APPRAISAL
5121300 EurekaRetail3327.4%1.47x6.0%5.8%1.28x451,182FAILNEG LEVERAGESTALE APPRAISAL
52Romeo CommonsRetail1877.4%1.45x5.5%1.7%1.08x1,433,835FAILNEG LEVERAGESTALE APPRAISAL
53Gravois Dillon PlazaRetail1365.5%1.07x0.8%(5.5%)0.76x163,178FAILNEG LEVERAGESTALE APPRAISAL
54MARKETPLACE AT FOUR CORNERS PHASE IIRetail1404.3%0.84x(1.8%)(9.6%)0.59x782,353FAILNEG LEVERAGESTALE APPRAISAL
55EAST TOWNE GREEN BAYRetail1547.0%1.39x4.6%2.6%1.13x704,674FAILNEG LEVERAGESTALE APPRAISAL
56PICK N SAVE APPLETON WIRetail1766.4%1.27x3.3%3.1%1.15x665,495FAILNEG LEVERAGESTALE APPRAISAL
ScreenAssumptionsSources
The computed columns on Compound's Screen sheet for all 31 deals: property, type and 12 of the columns to the right of the inputs, values as written. The three gate columns are left out and the Result column kept. Bold rows pass all three gates. Returns are model outputs at a price set to appraised value.

The gates and the rank by levered IRR

The shortlist ranks the passing deals by levered IRR and fills by formula, so it re-sorts when a shared assumption changes. It lists eight deals, from 16.8% down to -7.9%, and leaves ranks nine and ten blank.

Compound wrote the rank in column AO with a count of passing deals that have a higher IRR, and the shortlist looks each rank up with INDEX and MATCH. Change the spread, the hold period or a growth rate and the block reorders.

Midwest_CMBS_Deal_Pipeline_Screen.xlsx
C60fx=IFERROR(INDEX($C$26:$C$56,MATCH($B60,$AO$26:$AO$56,0)),"")
BCDEFGHIJKLMN
58TOP-TEN SHORTLIST: PASSING DEALS RANKED BY LEVERED IRR
59RankPropertyCityStateTypePurchase priceGoing-in cap rateYear-1 DSCRYear-1 cash-on-cashLevered IRREquity multipleNew loan less existing balanceFlags
601Bolingbrook Logistics CenterBolingbrookILIndustrial14,100,0009.4%1.89x11.8%16.8%1.93x3,392,198STALE APPRAISAL
612KINGSLEY APARTMENTSSterling HeightsMIApartment30,700,0009.7%1.94x11.4%14.5%1.77x(262,010)REFI SHORTFALL STALE APPRAISAL
623Clearpointe LandingMidlandMIApartment20,900,0008.6%1.72x9.3%13.5%1.73x185,000STALE APPRAISAL
634Colony Bay ApartmentsFort WayneINApartment25,000,00010.3%2.06x11.6%12.8%1.65x2,750,000STALE APPRAISAL
645Centennial HighlandsFort WayneINApartment18,550,0008.0%1.60x7.8%12.0%1.65x685,739STALE APPRAISAL
656WOODLAND GARDENS APARTMENTSRoyal OakMIApartment34,100,0008.6%1.72x8.8%11.8%1.62x(512,633)REFI SHORTFALL STALE APPRAISAL
667NORTHERN OHIO INDUSTRIAL PARKElyriaOHIndustrial15,900,00013.5%2.69x12.4%7.8%1.35x1,169,482STALE APPRAISAL
678LIBERTY CENTERTroyMIOffice33,000,00011.1%2.11x9.8%(7.9%)0.73x2,954,352STALE APPRAISAL
689
6910
ScreenAssumptionsSources
The top-ten shortlist block from Compound's Screen sheet, as written. Rows nine and ten are blank because eight deals pass. The Flags column joins the three flag columns for each deal.

Levered IRR of the eight passing deals

5-year hold, 65.0% loan at 6.90%, price at appraised value. Model output, in shortlist order

1. Bolingbrook Logistics Center16.8%2. Kingsley Apartments14.5%3. Clearpointe Landing13.5%4. Colony Bay Apartments12.8%5. Centennial Highlands12.0%6. Woodland Gardens Apartments11.8%7. Northern Ohio Industrial Park7.8%8. Office deal, MI-7.9%

Compound's Screen sheet, shortlist block, column K. Inputs: CMBS Form ABS-EE filings, Treasury par yield curve, BLS CPI, Zillow ZORI, Apartment List; growth by type, capital, spread, hold and exit cap spread are analyst assumptions.

Three rows deserve a second look, because a rank is a sort order and tells you nothing about why.

Bolingbrook Logistics Center ranks first. The filing reports 42,350 square feet at 100.0% occupancy. At the appraised value of $14,100,000 the price is $333 per square foot and the going-in cap rate is 9.4%. The appraisal is dated April 2017. The first question for the seller is what the property would appraise at today, because a 9.4% yield on a nine-year-old value says little about the yield on a current one. The second is the size. The filed revenue and net operating income come to $39.88 and $31.38 per square foot (our arithmetic on columns K, L and H), which is high for warehouse space. Check the square footage in the filing against the seller’s rent roll before trusting the rank.

Northern Ohio Industrial Park has the highest cap rate and ranks seventh. Its going-in cap rate is 13.5% and its coverage is 2.69x, the highest of the 31. Its IRR is 7.8%. The reason is the capital assumption. At $5 per square foot on 1,056,528 square feet, capital at purchase is $5,282,640, about 33% of the $15,900,000 price (our arithmetic). A flat capital figure per square foot is heavy on a property priced at $15 per square foot. Only a property condition report can say whether that fits this building.

The eighth-ranked deal passes every gate with a negative modeled return. It is an office building in Michigan. The filing reports 97.0% occupancy and the cap rate at appraised value is 11.1%, so it clears coverage at 2.11x. Its modeled IRR is -7.9%. The screen’s office assumptions produce that: 0.0% rent growth against 3.7% expense growth, and $25 per square foot of capital, which comes to $6,961,075 (our arithmetic). The gates test coverage, leverage and occupancy and never look at return, so a deal can pass all three with a negative IRR. Compound said in its answer that it would not carry this row forward on the pass alone.

Midwest_CMBS_Deal_Pipeline_Screen.xlsx
AY42fx=IF(AY$25>$D$15,0,$K42*(1+$V42)^AY$25-($K42-$L42)*(1+$D$18)^AY$25-$Y42+IF(AY$25=$D$15,$AD42-$AE42,0))
CZAFAGATAUAVAWAXAY
25PropertyTotal equityLevered IRREquity multipleYear 0Year 1Year 2Year 3Year 4Year 5
35Bolingbrook Logistics Center5,428,75016.8%1.93x(5,428,750)641,744679,971719,241759,5837,665,577
39NORTHERN OHIO INDUSTRIAL PARK11,165,6407.8%1.35x(11,165,640)1,383,3461,435,8771,489,4801,544,1699,180,616
42LIBERTY CENTER19,171,075(7.9%)0.73x(19,171,075)1,878,1481,781,3951,681,0261,576,9057,059,626
40Bingham Office Center32,361,300(41.9%)0.08x(32,361,300)371,482233,03089,403(59,592)2,065,423
41CONTINENTAL PLAZA5,802,825n/a-1.07x(5,802,825)(245,375)(281,193)(318,350)(356,896)(5,016,596)
ScreenAssumptionsSources
Levered cash flows for five of the 31 deals from Compound's Screen sheet: total equity, IRR, equity multiple and Year 0 to Year 5, values as written. Rows are reordered to put three passing deals first. Years 6 to 10 are zero at a 5-year hold and are left out.

The cash flow columns show how each IRR is built. Year 0 is the equity: price plus 2.0% closing costs plus capital, less the loan. Years 1 to 4 are income less debt service. Year 5 adds the sale, valued at the grown income divided by an exit cap rate 50 basis points above the deal’s own going-in rate, less 2.0% sale costs and the loan balance.

For the eighth-ranked office deal the annual cash flow falls from 1,878,148 in Year 1 to 1,576,905 in Year 4 as flat revenue meets rising expenses, and Year 5 brings 7,059,626 against 19,171,075 of equity. The two office rows below it do not pass the gates. The first, with 72.4% reported occupancy, shows a 0.08x equity multiple in the model after $13,121,300 of assumed capital (our arithmetic). The second, at 65.0% reported occupancy and a 2.8% cap rate at appraised value, has negative modeled cash flow in every year. Excel cannot solve an IRR for it, so the cell shows n/a and the equity multiple is negative.

All three office results rest on the same blanket inputs: an appraisal dated 2016 or 2017 used as the price, a new 6.90% loan, 0.0% rent growth, 3.7% expense growth and $25 per square foot of capital. Change any of them for a given building and its row changes. The figures are what the screen returns on those inputs and say nothing about how each building is doing under its existing loan.

The gates also exclude deals with better returns than some that pass. One apartment deal has a modeled IRR of 8.7%, above two deals on the shortlist, and is left off because its 7.0% cap rate at appraised value is below the 7.90% constant. Whether that is the right call depends on whether you buy for first-year yield or for total return. The screen applies the rule the request gave it.

Refinance shortfall and appraisal age flags

Five deals show a refinance shortfall: a new loan at 65.0% of appraised value would not pay off the existing balance. 30 of the 31 carry the stale appraisal flag, and the one exception is 901 West Madison, appraised in September 2025.

The refinance test asks a narrow question: at the shared terms, does the new loan cover the old one? Column AH is the new loan less the existing balance from the filing.

New loan less existing loan balance: the eight smallest of 31

New loan at 65.0% of appraised value, less the balance in the latest filing. Negative bars carry the refinance shortfall flag

Apartment deal D07, MI-$512,633Apartment deal D05, MI-$262,010Apartment deal D02, IN-$248,285Retail deal D21, IL-$114,424Retail deal D24, IL-$20,000Gravois Dillon Plaza, MO$163,178Clearpointe Landing, MI$185,000Novi Technology Center, MI$319,660

Compound's Screen sheet, column AH. Appraised values and loan balances: CMBS Form ABS-EE filings for the reporting period ending October 2026.

The five gaps are small next to the loans. The largest is $0.51 million against a balance of $22.68 million. The next two are $0.26 million against $20.22 million and $0.25 million against $11.43 million, and the last two are $0.11 million and $0.02 million. Of the five, two are apartment deals on the shortlist, at ranks two and six.

A shortfall here means the existing balance is more than 65.0% of an appraisal that is, in four of the five cases, more than three years old. It is a result of the screen’s assumed loan-to-value and says nothing about whether the existing loan will be repaid. Compound noted in its answer that the test uses the latest reported balance as a stand-in for the balance at maturity. An amortizing loan will be somewhat smaller by then, so the test leans conservative.

The appraisal age flag fires on anything older than three years at the analysis date. The ages in column AJ run from 1.2 years to 10.0 years. GRM Indianapolis, appraised in March 2023, is flagged at 3.6 years. With 30 of 31 flagged, the flag does not help you rank. It tells you that the price column of this screen is a placeholder for nearly every deal.

What the public data cannot tell you

The filings give you size, occupancy, income, an appraisal and the loan. They do not give you a price, a rent roll or the condition of the building, and those are what turn a screen into an offer.

Compound ended its answer with the list of what you would need from the seller. We agree with it and would ask in this order.

  • A price. A current appraisal or a broker’s opinion of value, and the seller’s expectation. Every return in this workbook moves with the price override in column Q.
  • Current income. A trailing-12-month statement and a current rent roll. Compound noted that the reported income runs through December 2025 or June 2026.
  • Leases. Abstracts and a rollover schedule for the office, retail and industrial properties. The screen grows revenue at one rate per type; a real building’s income steps when leases expire.
  • Capital needs. A property condition report and the capital history. The capital figures here are one number per property type.
  • The existing loan. Loan documents for the cost of early payoff or defeasance, the balloon amount and any extension options.
  • Taxes and insurance. A tax reassessment estimate at the purchase price and current insurance quotes. The screen grows all expenses at the consumer price index.
  • For apartments: in-place rents against market rents, recent lease trade-outs and concessions.
  • For the older industrial sites: a Phase I environmental report. Compound named Maple Street, built in 1919, GRM Indianapolis, built in 1940, and Northern Ohio Industrial Park, built in 1946.

There are also limits in the inputs themselves. The exit cap rate is each deal’s own going-in rate plus 50 basis points, so a high yield on an old appraisal carries through to the sale value. Occupancy is one figure per property on the filing’s date.

This is the handoff point to due diligence. The screen tells you which sellers to call first and what to ask them.

Our view: use the shortlist as a reading order

We would start with the six rows the screen ranks first and treat the last two as questions. The screen is good at removing deals; it is weaker at ranking the survivors, because the survivors are the ones with the oldest-looking prices.

A high cap rate at appraised value has two readings. The property may earn a lot for its price, or the appraisal may no longer describe the property. All eight shortlisted deals carry the stale appraisal flag. The screen cannot tell the two readings apart, and it says so with the flag.

What we would change before using this on a live pipeline:

  1. Type real asking prices into the override column. The gates then test the seller’s price against today’s loan terms, which is the question you are being paid to answer.
  2. Decide which negative leverage test you mean. Against the 7.90% constant, 23 deals fail. Against the 6.90% rate, 14 do (our count). Write the choice down next to the gate.
  3. Add a return floor to the Result column. The three gates let a deal through with a negative IRR. A fourth test on IRR or equity multiple would keep the eighth-ranked deal off the shortlist under these assumptions.
  4. Revisit office growth deal by deal. A 0.0% rate for six buildings in Michigan and Ohio is a starting point. A building with long leases to one occupier and a multi-tenant building with space to fill should not share it.

What would change our reading: a current appraisal on any shortlisted deal that comes in well below the filed one, or a lender quote wide of 175 basis points. Either moves the cap rate or the constant, and the shortlist with it.

A pipeline screen and a single-deal screen answer different questions. This one ranks many deals on thin data. The real estate deal screener page takes one teaser and tests it against a grid of prices and rents. The deals that survive both go to a full real estate pro forma, with a rent roll and a debt schedule in place of one growth rate.

Midwest CMBS deal pipeline screen (Excel)Compound’s workbook from this run: a one-sheet Screen with the shared assumptions block, 31 deal rows, gates, flags, a formula-driven top-ten shortlist and the levered cash flows; an Assumptions sheet with a source for every input; and a Sources sheet listing each filing.Download · Free

FAQ

What is real estate deal pipeline screening?

It is the first pass over a list of possible acquisitions, before any deal gets a full model. Each deal goes on one row with the same loan terms and growth assumptions, a few pass or fail tests, and a rank. In this workbook 31 deals share one assumptions block, three gates remove 23, and a formula-driven shortlist ranks the 8 that remain by levered IRR.

Which metrics belong in a deal screening model?

Price per unit or per square foot, going-in cap rate, year-1 debt service coverage, year-1 cash-on-cash return, levered IRR and equity multiple cover most first-pass questions. This screen adds a refinance test against the existing loan and three flags.

What is negative leverage in a real estate deal?

Leverage is negative when the property’s yield on price is below the cost of the debt, so borrowing reduces the cash return on equity. This workbook tests the going-in cap rate against the annual loan constant of 7.90%, which includes principal. Testing against the 6.90% interest rate is a looser version of the same idea.

Why use CMBS filings to build a deal pipeline?

CMBS trusts report each loan and property every month on Form ABS-EE: size, occupancy, revenue, net operating income, the appraised value and the loan’s balance, rate and maturity date. A loan near maturity has to be repaid or refinanced, which is why the request used maturity as the filter. The filings do not give an asking price, so this screen uses appraised value as a stand-in.

How do you compare apartment, industrial, office and retail deals in one screen?

Share the loan terms, hold period and gates, and vary only what differs by type: the size basis (units or square feet), rent growth and capital at purchase. This workbook keeps those in a four-row block by type.

How does the shortlist re-sort when an assumption changes?

The rank column counts how many passing deals have a higher IRR, and each shortlist cell uses INDEX and MATCH to find the deal with that rank. The IRR, the gates and the rank all point at the assumptions block, so one changed input reorders the shortlist.

CoatueCentri ConsultingAvra783 Capital PartnersOctahedron CapitalArix ResearchCoatueCentri ConsultingAvra783 Capital PartnersOctahedron CapitalArix ResearchCoatueCentri ConsultingAvra783 Capital PartnersOctahedron CapitalArix Research

See this workflow on your own deal

Try it now