Is there a free API for apartment REIT fundamentals? Delivered as one feed

"Fundamentals" for an apartment REIT is three different artifacts wearing one word, and a desk that buys only one discovers the other two missing at the worst possible moment.

The first artifact is the audited statement line: revenues, net income, total assets, long-term debt - filed under securities law and tagged so machines can read them. The second is the management-defined measure: funds from operations, same-store revenue and NOI, physical occupancy - constructs that exist because GAAP depreciation makes a leased building look like it is melting, and which print in earnings releases and supplemental packages rather than in standardized tags. The third is the market's running opinion: the multiple and the yield the equity trades at right now.

Datadory packages the whole stack. The catalog pools 14 datasets serving this industry - three primary records scoped to multi-family residential REITs, eleven cross-listed from neighboring slices - and one of those primaries scores a perfect 10 on our field-documentation rubric, against a cross-catalog mean of 7.81 across 1,744 datasets. The sections below walk the stack layer by layer, with the verified numbers each record ships.

The SEC EDGAR Company Search & Filing Archive is the spine: every disclosure the cohort files, joined with the standardized fact extract behind the documents. Verified against Equity Residential - trading as VMRK since the August 18, 2026 symbol change - the record holds 1,006 recent filings for the issuer and its operating partnership (CIK 0000906107), including eleven consecutive 10-Ks reaching back to 2015, the newest filed February 13, 2026 under accession 0001193125-26-051433. On the facts side, 593 us-gaap concepts arrive in a single pull of roughly 4.8 MB - see the XBRL company facts entry for how the structure works.

Three properties make it the right anchor for this cohort specifically. Every issuer tags against the same shared concept vocabulary, so one schema serves EQR, AVB, MAA, UDR, ESS and CPT without a bespoke parser per ticker. Every value keeps the accession number and form that reported it, so a restated quarter lands as another dated vintage instead of overwriting history - which turns point-in-time reconstruction into a filter rather than a rebuild. And the whole universe resolves through one CIK identifier reference map carrying 10,387 registered tickers, so assembling the comp set is a filter operation too, not six collection projects. Cross-sections ship pre-built: one view returns a single concept across all filers for a calendar period, and the CY2025Q1 Revenues cut alone held 1,934 facts.

Not from tags alone, and knowing where the tags stop is the difference between a model and a spreadsheet with holes. Funds from operations (FFO) and AFFO are management-defined constructs outside the us-gaap vocabulary, so they reach you two ways through Datadory: derived from the GAAP components the tagged feed already supplies, or captured straight from earnings-release exhibits via the same filing archive record.

Treat the differing counts as labeled views rather than an error: 15 in the FTSE overview, 18 cited as index-listed, 28 in the directory, because non-traded and recently listed constituents ride in the roster too. Read the label above each number before you cite it.

Deeper than the filings era, if the layers are chosen deliberately.

The Nareit REIT Market Data Reports Hub, cross-listed from Diversified Real Estate Activities, holds monthly FTSE Nareit index values and returns from December 1971 through June 2026 - 664 rows across 42 columns in the flagship workbook, indexed to exactly 100.0 on December 31, 1971. June 2026 All REITs closed at an index level of 11,717.261 on a 4.02% yield, and annual-grain studies get their own companion record spanning 1972 through 2025 (Nareit Annual Index Values & Returns). That is five decades of listed-real-estate returns sitting beneath fundamentals that mostly begin with the 2009 structured-reporting era, inside a document trail reaching the mid-1990s.

Four grains, arriving together - which is the whole integration problem in one glance:

# filing-grade fact row - Equity Residential / ERP Operating LP
cik        : 0000906107   entity : EQUITY RESIDENTIAL (VMRK since Aug 18, 2026)
form       : 10-K         filed : 2026-02-13    period : 2025-12-31
accession  : 0001193125-26-051433                document : eqr-20251231.htm
concept    : us-gaap:Revenues     val : 682329000     fy 2020  fp Q1  frame CY2020Q1

# market row - quote card, largest name in the cohort
ticker : VMRK   price : 63.66   mktcap : 24.61B
trailing_pe : 28.68   forward_pe : 47.85   fwd_dividend_and_yield : 2.81 (4.26%)

# demand row - CPS/HVS quarterly vacancy
year_quarter : 2026 Q2   rental_vacancy_us : 7.3   principal_cities : 8.0

Each row keeps its own unit, period and provenance stamp - millions here, dollars there, a market-close stamp on every quote cell - so the layers stack without lying to each other. That discipline is what lets one table hold a 10-Q's revenue figure, the sector's cash-flow conversion, the market's multiple on it and the vacancy rate behind next quarter's rents.

Ranked by how directly the stack answers the day job:

Start with the question, then take the smallest record that answers it. Statement-line fundamentals want the SEC EDGAR Company Search & Filing Archive; the sector denominator wants the Nareit Residential REIT Sector Page; multiples and yields want the quote-page record; the fifty-year return series wants the market data reports hub. For the filing ledger walked end to end, see the companion post on multifamily REIT filings as a managed feed.