What Investors Look for in a Developer

Revision notes#

This article was corrected on 27 September 2026. A check of its figures and attributions against the cited sources found errors, which have been corrected in the text. The full list, with what the article said and what it says now, is in the corrections section at the end.

This article was corrected on 29 September 2026. An earlier version described Victaura's positions, co-investment, investor protection, reporting and investment terms beyond what its public pages set out, and presented them as answers to the tests this article describes. Those descriptions were removed, and the text and the disclosure at the end now follow those pages.

Why the operator comes before the asset#

The asset is replaceable. The operator is not. A family office that allocates to private real estate is, in effect, underwriting a multi-year operational commitment by a counterparty it cannot replace mid-stream without absorbing material loss. The asset documentation can be re-negotiated. The capital structure can be refinanced. The brand can be swapped. The developer cannot be substituted once the structure is funded and the permit is in flight.

This is why institutional investors apply diligence to the operator first. Goldman Sachs' 2025 Family Office Investment Insights survey of 245 family offices, released on 10 September 2025, found that 44 per cent invest primarily directly in private real estate, while relying primarily on managers for other alternatives. When the family office is the underwriter, the operator is the variable under examination.

Five vectors anchor the work. Track record. Skin in the game. Structural protection. Alignment in the cap stack. Exit. They are the diligence map Victaura Research applies to any GP, drawn from the fund-level and statutory sources cited inline. What follows is an institutional reading of each vector, with source authorities cited inline.

Vector 1. Track record, completion ratio not announcement ratio

Launches are not deliveries. The first diligence error any LP makes is to count announcements. Press releases, master-plan renders, and ceremonial groundbreakings are not evidence of a developer's capacity to finish work. The relevant metric is the completion ratio: the percentage of announced projects that the operator has actually delivered, on the keys promised, in the configuration promised, within a defensible window of the timeline promised. Victaura Research treats verified delivery as the threshold question before any other diligence is conducted.

Three sub-metrics make the ratio readable. Delivered versus announced over a rolling ten-year window. Average time-to-completion against the original timeline. Number of projects abandoned, sold mid-stream, or transferred to a third-party completer. A developer with twenty announcements and four completed projects is a marketing operation. One with five announcements and five completed projects, smaller in headline value, is an operator.

Public registries make this verifiable. In Italy, the Comune issues and files every titolo edilizio and approved variant, and the property registers (Catasto and Conservatoria dei Registri Immobiliari) record transfers of registered title, not beneficial ownership. In Dubai, off-plan projects are registered with the Dubai Land Department, which requires a project escrow account for off-plan sales. In Indonesia, a transfer of land rights can be registered at the BPN (Badan Pertanahan Nasional) land office only on the basis of a deed made before a PPAT (Government Regulation 24/1997, Article 37). The diligence question is not whether the developer claims a track record. It is whether the registries confirm it.

Time-to-completion is the second filter. A project announced in 2020 for delivery in 2024 and delivered in 2024 is on schedule. A project announced in 2018 and still unbuilt in 2025 points to a permitting failure, a capitalisation failure, or both. The LP that compares announcement dates against delivery certificates and escrow-release milestones obtains a clean read on operational discipline. The press cycle does not provide this read. The registry does.

Vector 2. Skin in the game, the economics of GP co-investment

Co-investment is the most diagnostic single signal in the diligence file. A Summer 2025 practitioner guide in the PREA (Pension Real Estate Association) Quarterly states that real estate vehicle-level joint ventures typically follow a 90-10 or 95-5 split between LP and GP, a GP share of 5 to 10 per cent, and notes that some investors have shown flexibility around a 1 to 10 per cent GP commitment in joint ventures. Victaura Research treats 5 per cent as the working floor below which alignment becomes nominal; this is a Victaura Research assumption, not a published industry rule.

Below 5 per cent, Victaura Research reads alignment as weak. When a sponsor contributes less than 5 per cent of project equity while charging acquisition, asset management, construction supervision and disposition fees, the economics push the GP toward deal volume rather than long-term performance. A GP that wins on fee income whether or not the project performs is structurally indifferent to performance.

Fund-level and deal-level expectations diverge. At fund level, K&L Gates (January 2025) reports that the minimum sponsor commitment in a fund's LPA is often in the range of 1 to 5 per cent of aggregate commitments, consistent with diversification across a blind pool. At deal level, PREA describes a typical 90-10 or 95-5 LP-GP split in real estate joint ventures, and the case for a larger GP share is stronger because the GP is concentrating the family office's capital in a single named project. The LP is entitled to ask the GP to match that concentration with its own balance sheet.

The form of the co-investment is the third question. Cash at closing is the strongest form. Deferred or recycled fees treated as equity are weaker. Land contributed in kind at a valuation determined by the GP itself is the weakest. The diligence file should state form and timing, not just the headline percentage.

The same question applies to the publisher of this article. Victaura, through its parent Greystone B.V. (Netherlands), develops projects and holds a capital position in several of the markets discussed on this site, as the disclosure at the end of this article sets out. Those positions are a conflict for the reader to weigh. The form, timing and size of any commitment by Victaura are for the allocator to obtain in writing, as with any other developer.

Vector 3. Structural protection, the law before the marketing

The third vector is jurisdictional and statutory. Off-plan and development-stage real estate in 2025-2026 sits inside three principal protection regimes that any serious LP reads before it reads the brochure. Each regime carries an investor-protection mechanism that is enforceable against the developer in the local courts, and each carries a verifiable compliance trail.

Italy. Legislative Decree 122 of 2005. Article 2 of the decree obliges any developer selling property still to be constructed (immobili da costruire) to deliver the buyer a garanzia fideiussoria (bank or insurance surety bond) covering every sum collected from the buyer until the definitive transfer of title. Article 3 requires the surety to be issued by an entity authorised to conduct banking activities or by an insurance company authorised in surety operations. The Consiglio Nazionale del Notariato, in its Study 5813/C (2005), reads Article 2 as imposing a nullita' relativa on any contract not accompanied by the surety, which only the buyer can invoke. The protection is non-waivable. The buyer cannot, by contract, surrender it.

United Arab Emirates. Law No. 8 of 2007 (Dubai). The law applies to real estate development projects in the Emirate of Dubai (Article 3) and does not by itself govern projects in other emirates such as Ras Al Khaimah. Every off-plan project in Dubai is registered with the Dubai Land Department and sold through a project escrow account into which buyer payments are deposited. Under Article 14, the escrow agent retains 5 per cent of the value of the escrow account once the developer obtains the completion certificate, and releases it one year after the units are registered in the purchasers' names. Under Article 9, the developer's creditors cannot attach the payments deposited in the account.

Indonesia. The AJB and PPAT regime. Every sale of land rights in Indonesia, including sales to foreign-owned companies (PT PMA) and sales of Hak Pakai or Hak Guna Bangunan titles, requires execution of an Akta Jual Beli (deed of sale and purchase) before a Pejabat Pembuat Akta Tanah (PPAT, the licensed Land Deed Official), and other transfers by legal act, except auctions, likewise require a PPAT deed. Before drawing up the deed, the PPAT must check the land certificate against the records of the land office (Regulation of the Head of BPN 3/1997, Article 97) and whether the land is subject to a dispute. Under Government Regulation 24/1997, Article 37(1), a transfer of land rights by sale can be registered only if it is evidenced by a deed made before a competent PPAT. A purchase agreement that does not pass through a PPAT cannot be registered at the BPN, regardless of payment.

The institutional reading. A developer in any of the three jurisdictions either complies with these mechanisms or does not. The LP that asks for the fideiussione policy number, the RERA escrow agent reference, or the PPAT signing schedule, and receives a clean answer with documentation, is dealing with a compliant operator. Extra-jurisdictional structures designed to bypass these protections signal marketing over law. The choice of jurisdiction is itself a diligence vector.

Vector 4. Alignment in the cap stack, waterfalls and reporting

The fourth vector is contractual. Once track record, co-investment and statutory protection are confirmed, the LP turns to the document that governs cash flow. The distribution waterfall is where the developer either binds itself to LP outcomes or unwinds the alignment one tier at a time.

The institutional baseline. Limited partners receive return of capital first. They then receive a preferred return, which Wall Street Prep describes as usually 6 to 8 per cent on an annual basis, accrued on an IRR basis, before any promote accrues to the GP. Once the preferred is paid, a catch-up tier may apply, after which residual profit is split; the worked examples in EisnerAmper's February 2024 note use an 8 per cent preferred return and an 80 per cent LP, 20 per cent GP split after the catch-up. In a European waterfall, the sponsor receives carried interest only after the LPs recover capital plus the preferred return. In an American waterfall, the GP takes carried interest on a deal-by-deal basis before full return of LP capital; Victaura Research reads this as weaker protection for the LP in stressed scenarios, an assumption rather than a sourced finding.

The LP examines three sub-questions. Is the preferred return cumulative and compounded, or simple and lapsed? Is the catch-up 100 per cent to the GP up to parity, 50 per cent, or absent? Is the promote subject to a clawback in the event the final aggregate IRR falls below the preferred? Victaura Research reads a waterfall that omits any of these three protections as a negotiating signal; this is an assumption, not a sourced market standard.

Reporting cadence converts the waterfall into accountability. INREV, the European Association for Investors in Non-Listed Real Estate Vehicles, publishes the framework most institutional LPs anchor to. Its Guidelines comprise ten modules, including INREV NAV, performance measurement, fee and expense metrics, and reporting. The INREV NAV module provides a standard for calculating NAV, to improve transparency and comparability across different types of investment vehicles. Q&A guidance published by INREV in 2025 clarified how to adjust recorded goodwill for differences between IFRS values and economic value, and specified that deferred tax liabilities should be measured asset by asset rather than at a fixed percentage. A developer that publishes INREV-aligned quarterly LP letters, with INREV-aligned NAV and an annual audited account, is operating at the European institutional standard. A developer that publishes a one-page PDF of headline metrics is not.

ANREV in Asia Pacific and the NCREIF PREA Reporting Standards in North America are the counterpart frameworks, and INREV, ANREV, NCREIF and PREA work together towards global standards. The LP that operates across regions expects an operator to map its reporting to whichever framework dominates the LP's home market. The reporting standard is itself a proxy for operational maturity.

Vector 5. Exit, operator continuity versus flip-and-exit

The fifth vector is the one most developers prefer not to discuss. What is the operator's plan at handover? Three patterns dominate the branded residential and development-stage real estate markets in 2025-2026, and each carries a different risk profile for the LP and the end buyer.

Pattern one: developer holds the operating company. The developer retains long-term economic interest in the asset post-completion, through a residential management contract, a hotel operating company, or a ground-lease structure. The incentive to deliver to specification is reinforced by ongoing economics. This is the model of integrated developer-operators. Continuity of leadership and continuity of standards are protected because the operator is still in the asset.

Pattern two: developer hands off to an external brand operator. The developer builds, sells, and exits. A hotel brand (Marriott, Accor, Four Seasons, Hilton) takes over the residential management contract on a long-dated agreement. Marriott reports that since year-end 2023 it has grown its branded residential total portfolio by 23 per cent in Europe and 59 per cent in the Middle East and Africa (Marriott press release, 11 November 2025). Accor launched Accor One Living in January 2023 as a platform supporting the development, design and operation of mixed-use projects that often feature branded residences (Skift, December 2023). The brand carries the standard. Continuity becomes a function of the brand contract, not the developer.

Pattern three: flip and exit, no operator continuity. The developer completes, sells units, transfers common areas to a homeowners' association, and exits. It is not inherently a failure pattern, but it places the entire post-completion operational burden on the buyer collective. For LPs, the diligence question is whether the projected exit IRR depends on a brand or operator sale that has not yet been contracted. If yes, the deal carries exit risk that the underwriting model rarely scenarios properly.

The LP question is simple. What is the operator's economic position eighteen months after completion? A specific answer with named counterparties and signed term sheets is acceptable. A general answer about "brand partnerships" without documentation is not.

Red flags the institutional buyer walks away from#

Four warning signs follow from the five vectors. Victaura Research groups them as follows. Any one of them is a yellow flag. Two or more is the point at which an institutional buyer walks away regardless of the headline IRR.

First, no GP cash co-investment. The sponsor contributes nothing, or contributes land at a valuation it set itself, while charging acquisition, supervision and disposition fees. The economic interest is in transaction flow, not in performance. This is the single most diagnostic warning.

Second, opaque waterfall. The distribution mechanics are described in marketing materials in narrative form rather than in the LPA with worked numerical examples. The catch-up, the hurdles, the clawback, and the timing of preferred return accrual are left for negotiation "at closing". Institutional capital reads this as a deferred re-trade.

Third, jurisdiction shopping that bypasses statutory protection. The deal is structured through a holding entity in a third country specifically to avoid the buyer-protection mechanisms (fideiussione, RERA escrow, PPAT) of the asset's host jurisdiction. The structure is legally available but signals a preference for opacity over compliance.

Fourth, brand-only pitch. The investment thesis rests on a future agreement with a luxury hospitality or residential brand that has not been signed. The marketing deck shows the brand's logo or describes a "target partner". The LPA contains no covenant binding the developer to deliver the brand. The deal is, in substance, an option on a future negotiation the LP cannot enforce.

A fifth signal, less discussed, is opacity on prior litigation. A developer sued by previous LPs, buyers or lenders that does not disclose the dispute history is presenting a clean profile the public registry does not support. The LP that runs a court-record search before signing the LPA is being institutional, not paranoid.

What this means for the buyer, a diligence checklist#

For a family office, principal advisor or single buyer evaluating a private real estate sponsor in 2026, the diligence work resolves to a short, repeatable checklist. Each line is a document or a registry reference, not a conversation.

On track record: request the list of all announced projects in the last ten years with completion certificates, time-to-completion deltas, and the names of completed assets verifiable in the local Comune / DLD / BPN registry. Run independent checks.

On skin in the game: request the subscription documents showing GP cash co-investment at closing, in writing, with the form of contribution specified. Compare against the 90-10 or 95-5 LP-GP split that PREA describes as typical for real estate joint ventures. Treat anything below 5 per cent as a structural red flag, not a negotiating posture.

On structural protection: request the surety bond reference (fideiussione) for Italian deals, the RERA escrow account number for UAE deals, the PPAT signing schedule and certificate references for Indonesian deals. Treat the absence of these documents as the absence of compliance, not as an oversight.

On alignment in the cap stack: request the LPA, the waterfall worked example with numerical illustrations at multiple IRR scenarios, the reporting cadence schedule, and the NAV methodology. Map against INREV / ANREV / PREA where the LP's home market dictates. Require a clawback covenant.

On exit: request the operator continuity plan with named counterparties and signed term sheets, or a binding covenant to hold the operating company for a stated period. Treat brand logos in decks as marketing material until contracts exist.

Victaura is subject to the same checklist. Every line above is answered by a document or a registry reference, and a research article is no substitute for either. An allocator assessing Victaura should request the same documents from it and read the answers there, not in this article.

The closing point#

Five vectors. Five documents. Five answers. Track record verifiable in public registries, co-investment in cash, statutory protection compliant with host-jurisdiction law, alignment in the cap stack documented to INREV vocabulary, exit secured by signed term sheets. The developer that can produce these five answers in a diligence binder is the developer the institutional buyer can underwrite. The developer that produces a brochure instead is the developer the institutional buyer politely declines. The next decade favours documentation. The rules that rewarded opacity are narrowing: the UK non-dom regime was abolished from 6 April 2025, the CJEU ruled on 29 April 2025 that Malta's investor citizenship scheme is contrary to EU law, Spain's investor visa ended on 3 April 2025 under Ley Orgánica 1/2025, and DAC8, the EU rules based on the OECD CARF, applies from 1 January 2026. Victaura Research assumes that, as a result, buyers will increasingly select operators that publish constraints and disclose conflicts. The developer that hides its co-investment, its waterfall, its jurisdiction or its operator continuity is pitching a 2015 market. The buyer is no longer in it.

Skin in the game disclosure. Victaura, through its parent Greystone B.V. (Netherlands), develops projects on Lake Como (Italy), in Nungwi, Zanzibar (Tanzania), on Gili Air and in Uluwatu, Bali (Indonesia), and holds an off-plan capital position on Al Marjan Island, Ras Al Khaimah (UAE). Readers should assume commentary may be influenced by, or benefit, these positions. This document is classified as marketing material under MiFID II Article 24(3). It is not investment advice.

Corrections, 27 September 2026#

What the article said and what it says now. The published version attributed a 5 to 20 per cent deal-level GP co-investment norm to PREA and K&L Gates. Neither source contains it: PREA (Summer 2025) describes a typical 90-10 or 95-5 LP-GP split in real estate joint ventures, and K&L Gates (January 2025) reports a fund-level sponsor minimum that is often 1 to 5 per cent. The article now cites these figures and presents the 5 per cent threshold as a Victaura Research assumption. The published version credited the 44 per cent figure for direct investment in private real estate to the EY Family Office Guide; it comes from the Goldman Sachs 2025 Family Office Investment Insights survey of 245 family offices (10 September 2025). The published version said the Campden Wealth and RBC North America Family Office Report 2025 surveyed 360 family offices and ranked due diligence among the top three priorities; the report covers 141 North American family offices and contains no such ranking, so the sentence has been removed. The published version said Dubai Law 8/2007 retains a 5 per cent warranty fund for one year after completion; under Article 14, the 5 per cent is retained once the completion certificate is issued and released one year after the units are registered to purchasers, and the law applies to Dubai only. The published version described Marriott's 23 per cent branded-residence growth as EMEA-wide and dated it 2024; Marriott's release of 11 November 2025 gives 23 per cent for Europe and 59 per cent for the Middle East and Africa since year-end 2023. The published version gave a 70/30 split above a higher hurdle and a compounded preferred return with Wall Street Prep and EisnerAmper as sources; those sources give a 6 to 8 per cent preferred return on an IRR basis and 80/20 worked examples, and the 70/30 tier has been removed. Unsourced statements about a ULI threshold test, Ras Al Khaimah project dashboards, a 2024 Giustizia Insieme commentary, Campden roundtables and a 120 per cent completion benchmark have been removed or labelled as Victaura Research assumptions.

A further review on the same day refined some of the corrections above and fixed further points. The article said that the INREV NAV module sets out adjustments for reconciling IFRS balance sheet values with economic NAV, and that INREV guidance on goodwill and deferred tax was published in October 2025. The INREV standards page describes the module only as a standard for calculating NAV, and INREV's page of 13 October 2025 says the goodwill and deferred tax Q&A guidance was published in 2025 without giving a month, so the article now says so. The article said that every transfer of real property in Indonesia requires an Akta Jual Beli. Under Government Regulation 24/1997, Article 37(1), as set out in the LITRA Jurnal Hukum study of October 2023, the AJB is the deed for a sale, other transfers by legal act need a PPAT deed, and auctions are the exception; the article now says this. The article described a waterfall without a cumulative preferred return, catch-up terms or a clawback as 'not market'. No cited source supports a market standard on this point, so the statement is now labelled as a Victaura Research assumption.