ReiForge
ReiForge
Loading workspace…
Your book — aggregated metrics, risk, and a two-deal compare under one assumption set. Data export and Labs stay collapsed until you need them.
⚠ Outcome lookup isn’t wired yet: ReiForge doesn’t currently link a passed deal’s address to a CREVault property record, so no comps/AVM data on what actually happened is shown here — regardless of whether CREVault is keyed. This is a known gap (see P2-5), not a missing value being hidden.
No passed deals logged yet. Decisions you log as “Pass” on the underwrite page will show up here so you can look back on them.
Your portfolio is empty. Search real listings in Discover, import a CSV above, or Load sample data to explore with 25 fictional Atlanta deals.
Dormant — your deals live only in this browser. Add Supabase keys (NEXT_PUBLIC_SUPABASE_URL + NEXT_PUBLIC_SUPABASE_ANON_KEY) to sign in and sync deals across devices.
The same portfolio, re-underwritten under each assumption set. Year-1 metrics and the cash-on-cash accuracy below both use vacancy & expense; rent growth is versioned with the set but does not yet affect the figures shown here.
Create or edit a set
Saving a name that already exists adds a new version of that set (its own track record starts fresh) — the built-in three can’t be overwritten this way.
| Metric | Conservative | Base | Aggressive |
|---|---|---|---|
| Vacancy | 8% | 5% | 3% |
| Expense ratio | 50% | 40% | 35% |
| Rent growth | 2.0% | 3.0% | 5.0% |
| Portfolio NOI (yr 1) | $0 | $0 | $0 |
| Cap rate (on cost) | — | — | — |
| Cash-on-cash | — | — | — |
| DSCR | — | — | — |
Record realized cash-on-cash outcomes (Realized outcomes card) to rank these sets by how well each one predicts reality.
Your own targets and preferences — used to default the Discover search and buy-box targets, and shown to the AI features as context (it never overrides an advisory or a computed metric).
No profile set yet — every default below falls back to the app’s built-in values.
Edit profile
Blank fields are left unchanged — this only updates the fields you fill in.
Record whether a deal actually worked out. We compare the model’s probability against reality, fit a recalibration map, and apply it.
Record at least one realized outcome to measure calibration.