AI Deal Analysis: What It Reviews for Each Deal Type

What the AI actually does when you paste a deal

Most deal calculators ask you to fill in the ARV and spit back a MAO. That's not underwriting — that's arithmetic. What DealDog's AI does is closer to what a seasoned operator does when they pull up a deal at the kitchen table: it reads the full picture, flags the risk, and tells you which exit makes the most sense given the numbers you fed it.

Paste in a deal description or fill the intake form at calculator.dealdogcrm.com and the AI returns a full analysis in roughly 60 seconds. That output includes ARV range, repair estimate range, MAO by exit strategy, projected cash flow or assignment margin, risk flags, and a ranked recommendation on how to exit the deal.

The specific feedback varies by deal type — and that matters a lot. A SubTo deal doesn't need the same analysis as a mobile home park acquisition. A novation has different risk flags than a straight assignment. The sections below break down exactly what the AI examines and what it tells you for each of the 15 deal types the platform underwrites.

Wholesale assignments and what the AI flags beyond MAO

Wholesale is where most operators start, but the analysis shouldn't stop at "ARV minus repairs minus margin equals MAO." The AI looks at spread viability — specifically whether the assignment fee you're targeting is realistic given comparable dispositions in that submarket, not just whether the math clears on paper.

For a standard wholesale assignment, the feedback covers:

• ARV range with a confidence flag (thin comps vs. strong comp pool)
• Repair estimate range by condition tier (light lipstick, moderate, full gut)
• MAO at multiple assignment fee targets, not just one number
• Marketability score — how fast this deal type and price point typically moves in the area
• Risk flags: title issues indicated in the description, deferred maintenance keywords, flood zone language, non-arms-length sale indicators
• Buyer match readiness: whether the deal profile matches active buyers in the DealDog network before you even lock it up

The buyer match readiness piece is something most calculators skip entirely. There's no point locking up a deal at a number that clears your MAO if your buyer list can't absorb it. The AI flags that disconnect before you're under contract.

Operators I know have used this specifically for the thin-comp problem. When you're working a rural county or a secondary market where Zillow has three sales in 18 months, the AI surfaces the confidence level on its own ARV estimate rather than returning a false precision number. That honest flag is worth more than a confident wrong answer.

Fix-and-flip analysis: where the AI earns its keep on the cost side

Flips live and die on the gap between your all-in cost and net sale price. The ARV piece is table stakes. Where operators consistently bleed is on carry, financing, and scope creep — and that's where the AI's feedback gets specific.

For a fix-and-flip deal, the analysis returns:

• ARV with comp confidence rating
• Repair estimate broken into scope tiers (cosmetic, structural, systems-level)
• Estimated hold period based on market velocity data for that asset type and price band
• Financing cost range (hard money or private money assumptions built in, adjustable)
• Net profit projection with a conservative, base, and optimistic scenario
• Risk flags: over-improvement risk (ARV ceiling vs. scope investment), permit-trigger language, foundation or roof keywords in the description
• Exit alternative: if the flip margin is thin, the AI surfaces whether a wholesale or novation exit outperforms on a risk-adjusted basis

That last point is one operators miss. Plenty of deals come in that look like flips but underwrite better as assignments once you factor in hard money carry and a 5-month hold. The AI surfaces that alternative rather than forcing you to run the same deal twice in two different calculators.

As of Q2 2025, hard money rates at most direct lenders are running in ranges that make carry cost one of the most sensitive variables in a flip model. A deal that looks like a $40k profit on a 3-month hold can compress significantly on a 6-month hold with a lender charging origination plus monthly interest. The AI stress-tests that hold period automatically.

BRRRR deals: what the refi leg looks like in the AI output

BRRRR is a two-act deal and most calculators only model the first act. The AI underwrites both: the acquisition and renovation phase and the refinance and hold phase, with the connection between them being the ARV-to-refi-value relationship.

The BRRRR analysis output includes:

• Post-rehab ARV
• Estimated stabilized value for refi purposes (which sometimes differs from retail ARV depending on lender treatment of the asset class)
• Projected refi proceeds at 70-75% LTV (the conventional DSCR refi range as of Q2 2025)
• Cash left in deal after refi — the number BRRRR operators actually care about
• Monthly cash flow projection at market rent
• DSCR at projected rent vs. projected PITIA payment
• Risk flags: equity strip risk (if ARV is thin relative to rehab cost), rent-to-value ratio warning, market rent confidence flag

The DSCR flag is worth highlighting separately. DSCR lenders — institutions like Kiavi, Visio Lending, or your local portfolio lender — underwrite to a minimum ratio, typically 1.0 to 1.25 depending on the program. If the AI sees a projected rent-to-PITIA ratio that falls below 1.0, it flags that the refi exit may require a cash injection rather than returning capital. That's a deal-structure problem, not a repair-estimate problem, and catching it before you're 90 days into a rehab saves real money.

SubTo and seller finance deals: what the AI actually models here

Creative finance deals are where generic calculators completely fall apart. The AI handles SubTo, seller carry, wraparound mortgages, and hybrid structures because the underwriting logic is fundamentally different from a cash or hard-money deal.

For a Subject-To acquisition, the analysis returns:

• Existing loan balance and remaining term (based on what you input)
• Payment stack: PITI on the existing loan plus any seller carry second
• Monthly cash flow projection at market rent
• Equity position at acquisition vs. projected equity at disposition
• Due-on-sale risk flag based on loan type (FHA, VA, and conventional behave differently)
• Exit strategy ranking: hold for cash flow, retail sale via novation, wholesale to another investor

For seller finance and wrap deals, the AI models the spread between the underlying financing cost (if any) and the seller carry terms, and projects cash-on-cash return at the terms you negotiated. It also flags amortization mismatches — situations where a short balloon term on the seller carry creates a refinance risk before the deal has had time to appreciate or season for conventional financing.

Operators running high-volume SubTo deals have told me the due-on-sale flag alone is worth the tool. It doesn't make the decision for you, but it surfaces the flag with context about the loan type so you're walking into the deal informed rather than finding out about VA loan servicing rules from a title company 60 days later.

The DealDog platform also handles Morby Method (a hybrid seller carry plus SubTo structure) as a named deal type, which is something even most creative finance-specific tools don't recognize as a distinct underwriting category.

Novation, lease option, and the exit strategies most tools skip

Novation assignments and lease options are two of the most underserved deal types in the calculator market. Most tools either don't recognize them or lump them into a generic "creative finance" bucket with no structure-specific modeling.

For a novation, the AI models:

• Projected net sale price at retail ARV
• Agent commission assumption (since novations typically go on MLS)
• Closing cost range
• Wholesale fee or profit split structure
• Holding cost during the listing period
• Risk flag: days-on-market sensitivity (the deal's profitability drops materially if it sits longer than projected, and the AI shows that curve)

For a lease option, the output covers:

• Option consideration received
• Monthly cash flow during the lease period
• Projected equity at option exercise
• Risk flag: option expiration without exercise (what you're left holding and at what basis)
• Exit alternative if the tenant-buyer doesn't exercise: rental hold, retail sale, or wholesale assignment of the option itself

The days-on-market sensitivity flag on novation deals is something I haven't seen any other calculator surface. A novation that pencils at 30 days on market often doesn't pencil at 90. The AI shows you that inflection point so you can decide whether the deal needs a price reduction built into the model before you commit to the structure.

Commercial, multifamily, land, MHP, and storage: what changes in the output

Commercial deal types require income-based underwriting, not just ARV-based analysis. The AI switches frameworks automatically based on the deal type you select.

For multifamily (2-4 units through small apartment buildings), the output includes:

• Gross rental income at market rents
• Vacancy and credit loss assumption
• Operating expense ratio by property class
• Net operating income (NOI)
• Cap rate at acquisition vs. market cap rate for the subtype and area
• DSCR projection
• Value-add scenario: projected NOI post-stabilization and what that does to value at the same cap rate

For mobile home parks (MHP), the model treats lot rent separately from home income, flags infrastructure ownership (city water/sewer vs. private well/septic), and adjusts cap rate assumptions accordingly. MHP deals that look like screaming values at face value often carry infrastructure liability that materially changes NOI — the AI flags that based on keywords in your deal description.

For land, the analysis pivots entirely: no rental income, no ARV in the traditional sense. The output covers entitlement status, comparable sales per acre or per lot, subdivision potential flag, and carry cost projection during the hold period before sale or development.

For self-storage, the AI runs a unit-mix model — revenue per unit type, stabilized occupancy assumption, and NOI — and compares the implied value to replacement cost per square foot, which is how institutional storage buyers actually underwrite.

According to the SBA's small business economic indicators and commercial lending data tracked by the Federal Reserve's Z.1 Financial Accounts, commercial real estate financing conditions have tightened meaningfully since 2022, making pre-offer underwriting on income-producing assets more important than it was in the 2020-2021 environment when deals were moving faster than analysis could keep up.

The one thing the AI does that no calculator does: ranked exit strategy output

Here's something that contradicts how most operators think about deal analysis tools: you shouldn't be picking your exit strategy before you run the numbers. The market should tell you the exit, not your preference.

Most calculators work backwards from a fixed exit. You open the "flip calculator" and run flip numbers. You open the "rental calculator" and run rental numbers. The deal type is a field you fill in before the analysis, not a conclusion the analysis reaches.

DealDog's AI returns a ranked exit strategy output for every deal. Feed it the same deal and it simultaneously models wholesale, flip, BRRRR, rental hold, novation, and SubTo (where applicable given the deal's structure) and ranks them by projected return, risk profile, and speed-to-liquidity. Then it tells you which one it recommends and why.

A SubTo deal that most operators would hold for cash flow might actually rank highest as a novation if the ARV is strong and the existing loan balance creates enough equity to retail it. A flip deal might underwrite better as a wholesale assignment if the repair scope triggers structural work that pushes carry costs into a range where the margin disappears. The AI surfaces those findings without you having to run the same property through six different calculators.

Marcus, a multifamily operator working the Midwest secondary markets, ran a 6-unit deal through the analyzer expecting a straightforward BRRRR recommendation. The AI ranked a value-add hold above the BRRRR exit because the projected refi proceeds at 70% LTV left too much cash in the deal relative to the projected cash-on-cash return in stabilized hold mode. He restructured the offer price based on that output and negotiated seller carry to reduce the capital requirement. The deal closed two weeks later with better cash-on-cash than his original model projected.

Before your next deal closes: the 3-step analysis workflow

Running a deal through the AI isn't a one-time pre-offer check. The operators getting the most out of it run the analysis at three points in the deal timeline.

1. Before the offer: Run the deal description through the analyzer to get the ranked exit output and MAO range. Use the MAO to set your offer floor, not your target price. If the AI flags thin comps or a risk issue, factor that into the negotiation position.
2. After the inspection or walkthrough: Update the repair estimate inputs with actual contractor numbers and re-run. The first analysis uses range estimates. The second run with real scope gives you the accurate MAO and updated exit ranking. If the deal no longer pencils at the agreed price, you have documented basis for a price reduction request.
3. Before disposition: Run the deal one more time with the updated ARV from your appraisal or BPO. Check whether the buyer match output has shifted since you locked it up. Markets move. The buyer who was the best match at lockup may not be the best match at disposition if their buy box has changed. DealDog's cross-network matching re-scans at disposition, not just at intake.

If you want to see the full analysis output for your current deal type, the deal flow calculator is live at calculator.dealdogcrm.com — no account required for the first two analyses. If you're ready to run unlimited deals across all 15 types and get cross-network buyer matching built in, the Core tier is free forever. See everything the platform does at dealdogcrm.com.

Frequently Asked Questions

What deal types does DealDog's AI analyze?

DealDog underwrites 15 deal types: wholesale assignment, fix-and-flip, BRRRR, rental hold, SubTo (Subject-To), seller finance, wraparound mortgage, novation, lease option, Morby Method, multifamily, land, mobile home park, self-storage, and mixed-use commercial.

Each deal type uses a different underwriting framework. Commercial and income-producing assets run on NOI and cap rate logic. Residential acquisitions run on ARV and repair estimates. Creative finance deals model the payment stack and equity position at various exit points.

What feedback does the AI give you after analyzing a deal?

The output includes ARV range with a confidence rating, repair estimate range, MAO at multiple fee targets, projected cash flow or margin, risk flags pulled from your deal description, and a ranked exit strategy recommendation showing which exit (flip, wholesale, BRRRR, rental, novation, etc.) returns the best outcome given the current inputs.

Risk flags cover things like thin comp pools, over-improvement risk, DSCR compression on BRRRR exits, due-on-sale exposure on SubTo deals, and days-on-market sensitivity on novations. These aren't generic warnings — they're triggered by specific inputs or keywords in your deal description.

How accurate is the AI's ARV estimate?

The AI returns an ARV range, not a single number, and flags the confidence level based on comp density in the submarket. In areas with thin sales data — rural counties, secondary markets with few recent transactions — the output will explicitly flag low confidence rather than returning a false precision number.

For high-confidence ARV, you still want a BPO or appraisal before you close. The AI's ARV is best used as a pre-offer sanity check and negotiation anchor, not a final underwriting number for a lender or hard money draw schedule.

Can the AI handle creative finance deals like SubTo and seller carry?

Yes. SubTo, seller finance, wrap mortgages, lease options, and Morby Method are all named deal types with structure-specific modeling. For SubTo deals, the AI models the existing loan's payment stack, flags due-on-sale risk by loan type, and ranks hold vs. novation vs. wholesale exit by projected return.

This is where generic calculators break down. A SubTo deal with an FHA loan in the existing stack carries different risk than one with a conventional loan, and the AI surfaces that distinction based on what you input.

Does the AI match deals to buyers automatically?

Yes. After the analysis runs, DealDog scans every buyer in the network against the deal's profile — location, price range, asset type, condition, and exit strategy — and surfaces qualified matches. On Core+ and Pro, this scan covers the entire DealDog cross-network, not just buyers registered to your account.

JV referral fees are built into the matching workflow, so if a deal matches a buyer registered under a different operator's account, the fee split is structured automatically rather than requiring a separate side agreement.

How is this different from a standard MAO calculator?

A MAO calculator takes your ARV and runs one formula. DealDog's AI reads your full deal description, identifies the asset class and applicable deal structures, models multiple exits simultaneously, and returns ranked recommendations with risk flags specific to that deal's profile.

The practical difference is that a MAO calculator confirms math you already did. The AI catches things you didn't model — over-improvement risk, DSCR compression on the refi leg, days-on-market sensitivity on a novation, infrastructure liability on an MHP. Those are the items that kill deals after you're already under contract.