Canonical page: https://sequoia-research.com/writing/top-dst-sponsors-2026-why-different-ranking-methods-produce-different-leaders
Author: Sequoia Research
Published: 2026-10-18T03:09:38.165Z
Updated: 2026-10-18T03:09:38.165Z

# Top DST Sponsors (2026): Why Different Ranking Methods Produce Different Leaders

## What Is This Article About?

A search for top DST sponsors can lead to lists that disagree even when they use the same facts. This article shows how that happens through a small study of three public records reviewed in 2026. We change the weights and decision rules while keeping the observed document features fixed. The results rank those selected records for narrow research tasks; they do not rate the sponsors or their investments.

## Table of Contents

- [Why Top DST Sponsor Lists Disagree](#why-top-dst-sponsor-lists-disagree)
- [The Public Records Used in This 2026 Study](#the-public-records-used-in-this-2026-study)
- [How the Document Features Were Coded](#how-the-document-features-were-coded)
- [Different Weights, Different Research Leaders](#different-weights-different-research-leaders)
- [Equal Weights and Hidden Double Counting](#equal-weights-and-hidden-double-counting)
- [Ties, Rounding, and Ranking Sensitivity](#ties-rounding-and-ranking-sensitivity)
- [Hard Requirements Versus Weighted Scores](#hard-requirements-versus-weighted-scores)
- [Missing Data and Sponsor Ranking Limits](#missing-data-and-sponsor-ranking-limits)
- [How to Read a DST Sponsor Ranking Methodology](#how-to-read-a-dst-sponsor-ranking-methodology)
- [Write a Research Decision That Others Can Recheck](#write-a-research-decision-that-others-can-recheck)
- [FAQ](#faq)

## Why Top DST Sponsor Lists Disagree

The word top hides a choice. One reader wants a source that explains reporting practices. Another wants a file with past outcome rows. A third wants help finding which programs belong in a review. Those are separate research needs. A single list can serve one need well and answer another poorly.

Start by asking what the list ranks. It might rank firms, individual offerings, public records, service experiences, or search results. These objects are related, but they are not the same. A clear public record may help you begin research. It does not establish the quality of a particular property or the soundness of its debt.

This article uses a narrow object: the content of three selected public records. That choice lets the reader inspect what earned each point. It also places a firm boundary around the result. A point means that a defined feature appears in the record. It says nothing about whether an investment will perform well.

Think of the score as a way to choose which document to read first for a stated task. If you need a policy statement, a record containing that statement may come first. If you need rows for a later outcome review, a different record may come first. The order can change without either document changing.

This is why a ranking needs more than a list of names and numbers. The reader needs the question, the source set, the rules, and the weights. Without them, a score of 80 has no clear meaning. It could represent a useful research fit, a broad opinion, or an unstated mix of both.

A Delaware statutory trust, or DST, is an ownership structure used in some real estate investments. Our document exercise does not test any trust's legal status or tax treatment. It does not measure financial strength, investor returns, fees, or the chance of loss. Those topics need their own evidence and methods.

Nor is this a market survey of every sponsor. We chose three accessible records to make the effect of method choices visible. Selection can shape a result before any arithmetic begins. Adding a record, using a different page, or defining a feature more broadly could change the table. A reader should see that limitation beside the score.

## The Public Records Used in This 2026 Study

We assembled this small source set on October 17, 2026. The records cover different periods and serve different purposes. The date of our review does not turn them into a common-period financial dataset. We use short document labels below to keep the analysis tied to exact sources.

**Record A — Capital Square's real estate investment page.** The page states an annual independent cash-flow audit practice for its DST properties. We inspected the page, not the underlying audit reports. This selected page does not contain individual full-cycle outcome rows or a numeric count of a defined program population. Its links to other material are outside this record's boundary. We call it Record A throughout the calculations. [Capital Square source page](https://capitalsq.com/real-estate-investment/).

**Record B — ExchangeRight's investment track-record PDF.** The eight-page file is dated August 31, 2026. It includes individual full-cycle rows and defined program counts. Its summary lists 103 active and 34 full-cycle offerings across several vehicle types; the 34 are not all DSTs. We did not find the narrowly defined annual property cash-flow audit policy in this file. Valuation references do not satisfy that different feature. We call this file Record B. Its return notes also distinguish fee structures and certain investor exchange outcomes, so we do not treat the rows as uniform cash-sale results. [ExchangeRight source PDF](https://www.exchangeright.com/resources/investment-track-record.pdf).

**Record C — Inland's portfolio overview landing page.** The page describes an overview of 334 programs offered through December 31, 2025. We inspected the landing page; the full underlying report was not obtained for this exercise. The page itself does not show individual full-cycle outcome rows or the narrowly defined audit policy. The count is not presented here as 334 completed exits. We call this selected page Record C. [Inland source page](https://inland-investments.com/insights/private-investments-portfolio-overview).

These are company-published statements. We have not independently audited their underlying records. Our original work is to code the presence of three features and test how declared rules order that fixed table. The study measures neither the truth of every underlying claim nor the completeness of each firm's entire research library.

The distinction between a page and a firm matters most when a feature is absent. A firm may have relevant material elsewhere. We did not search every possible private file, investor portal, or related entity's archive. A zero in this study cannot support a claim that a sponsor lacks the practice or experience in question.

The three records are also different lengths. A short landing page and a full PDF have different space and purposes. That can affect feature presence. We preserve this mismatch because it is part of the lesson: a neat score can conceal an uneven source selection. The result is useful for this document exercise, not a fair overall contest among firms.

## How the Document Features Were Coded

We use three yes-or-no features. Each has a narrow written test. A one means the feature was observed in the inspected record. A zero means it was not observed within that record's stated boundary. Neither value is a sponsor quality grade.

| Feature | What Earns a One | What Does Not Establish It |
|---|---|---|
| P: Policy statement | An explicit statement of an annual independent property cash-flow audit practice | A general promise of transparency, a valuation reference, or a report title alone |
| R: Outcome rows | Individual full-cycle outcome rows in the inspected record | A link to a separate file that was not inspected as part of that record |
| C: Population count | A numeric count tied to a defined program or offering population | A vague claim of broad experience or an unrelated count of people |

The count feature records presence, not size. It does not award more points for a larger count. Program definitions differ, and a bigger number does not by itself establish better results. We also do not score the reported returns in the outcome rows. Their presence creates a starting point for further work, not a validated performance comparison.

The coded dataset is deliberately small:

| Selected Record | P | R | C |
|---|---:|---:|---:|
| Record A | 1 | 0 | 0 |
| Record B | 0 | 1 | 1 |
| Record C | 0 | 0 | 1 |

Keep the feature tests next to this table. If they are removed, the zeros could look like failed audits or poor returns. They mean neither. The table answers only whether the specified content appears in the selected record. That narrow meaning must travel with every chart or excerpt based on it.

A reviewer should be able to repeat each coding decision. Save the source address, review date, record boundary, and relevant page or section. Write down why close cases did not qualify. For example, two kinds of third-party work can sound similar but have different subjects. The test should follow its stated subject, not a loose impression of credibility.

There is a separate state for an unreadable record. If a page cannot be opened, its feature values are unknown. You cannot turn failed access into three zeros. In this study, the selected records were inspected; the linked report outside Record C was not part of the coded record. Changing that boundary would require a new review.

This coding system is an educational choice. It has not been validated as a measure of sponsor quality. The letters and arithmetic make the method easy to check. They do not make the chosen features complete, equally useful, or sufficient for an investment decision.

## Different Weights, Different Research Leaders

A weight is the importance assigned to a feature. For this exercise, weights add to 100. Multiply each feature's zero or one by its weight, then add the results. The score therefore runs from zero to 100. It is a document-fit score under the stated method, not a percentage chance of success.

### Method One: Start With the Reporting Policy

Suppose the research task is to find a public statement about annual property cash-flow audits. Give P a weight of 60, R a weight of 20, and C a weight of 20. Those choices place the policy feature ahead of the other two combined.

Record A earns 60 points: 60 times one, plus two zero terms. Record B earns 40: zero for P, 20 for R, and 20 for C. Record C earns 20 for C. The order is A, B, C. This is a useful reading order for the chosen policy question, subject to the limits already described.

It would be wrong to relabel these results as sponsor safety scores. A policy statement does not equal a completed inspection of the actual audit work. And the other records' zeros do not show that the firms lack audits. The arithmetic is exact while the meaning remains narrow.

### Method Two: Start With Individual Outcome Rows

Now suppose the first task is to locate rows for a later full-cycle analysis. Keep the same dataset, but assign P 20, R 60, and C 20. Record A scores 20. Record B scores 80. Record C scores 20. B leads, and A and C tie behind it.

Nothing in the source table changed. No new result appeared, and no company improved or declined. The research question changed the weights. That is enough to produce a different leader.

The rows would still need careful review before they supported a performance claim. A presence test does not normalize dates, fee classes, invested amounts, or exit methods. It only tells us where the selected content can be found. The next task may require a very different comparison.

| Method | P Weight | R Weight | C Weight | A Score | B Score | C Score |
|---|---:|---:|---:|---:|---:|---:|
| Policy first | 60 | 20 | 20 | 60 | 40 | 20 |
| Outcome rows first | 20 | 60 | 20 | 20 | 80 | 20 |

The words above the numbers carry much of the meaning. If a chart retained only the last three columns, a reader could mistake a change in priorities for a change in the firms. A responsible comparison shows the method name, weights, feature definitions, and source date with the score.

Do not choose the weights after deciding which name should lead. Write the research purpose first. Then explain why each feature serves that purpose. If a feature has little bearing on the question, give it a lower weight or remove it with a stated reason. Record the choice before looking at the resulting order when possible.

A preference is still a preference even when expressed as a precise number. There is no evidence in this exercise that a policy statement is exactly three times as useful as an outcome-row feature. The 60-to-20 ratio is a declared scenario. It tests what follows from that choice; it does not prove the choice is objectively correct.

## Equal Weights and Hidden Double Counting

Equal weights can feel neutral. They are easy to explain: each feature receives one-third of the total weight. In this dataset, A and C each have one feature, while B has two. Their scores are therefore about 33.33, 66.67, and 33.33. The exact fractions, rather than rounded displays, establish the tie.

But equal weighting begins after someone chooses the features. That earlier choice can create a strong preference. A list with five measures of past outcomes and one measure of reporting policy already gives more room to outcomes. Assigning equal weight to each line does not undo that design.

Consider an original variation on our method. Keep P and C, but split R into two labels that both record the same outcome-row presence. Now there are four equal features. B earns three of four points; A and C each earn one. The scores become 75, 25, and 25, respectively. No new source evidence has been added.

This is hidden double counting. The total is mathematically correct, but the method has quietly given the repeated feature more influence. Renaming the second line does not create another independent fact. A careful reader should ask whether two columns really measure separate things or merely describe the same observation.

The problem can also arise with more complex sponsor data. Total offerings, total assets, and total properties may overlap as measures of business scale. They are not identical, but treating each as a fully separate quality signal needs a reason. Otherwise, a large firm could receive several rewards for much the same underlying size.

A practical repair is to group related measures before assigning weight. For example, give a broad research category one budget of points. Then divide that budget among its subfeatures. Adding a useful subfeature within the category need not increase the category's total influence. This makes the tradeoff visible.

The repair does not make the weights objective. It makes the choice easier to inspect. Readers can see how much of the total comes from each broad question. They can also see whether an added line supplies new evidence or mainly changes the amount of weight placed on an existing idea.

Avoid interpreting ordinal grades as exact distances without a stated rule. A label of excellent is not automatically twice as good as fair. If a publisher turns words into numbers, ask how the scale was defined and applied. The arithmetic can look precise even when the starting judgments have wide room for disagreement.

## Ties, Rounding, and Ranking Sensitivity

A ranking is sensitive when small method changes alter the order. Our fixed table makes the switching point easy to see. Let the policy weight be p. The other two weights together equal 100 minus p. A's score is p, while B's score is 100 minus p.

They tie when p equals 50. A leads when the policy weight is above 50. B leads when it is below 50. This conclusion holds for any nonnegative split of the remaining weight between R and C. It follows from the table's shape, not from a new claim about either sponsor.

Set the weights to 50 for P, 30 for R, and 20 for C. The scores are A 50, B 50, and C 20. Using competition ranks, A and B share first place and C is third. Competition ranks count how many entries have strictly higher scores, then add one. A different tie convention should be named if used.

Now shift just one-tenth of a point. With weights 49.9, 30.1, and 20, A scores 49.9 and B scores 50.1. Reverse that shift to 50.1, 29.9, and 20, and the order reverses. C remains at 20. Both leading scores could display as 50 if rounded to whole numbers.

That display would hide the difference rather than create a true tie. Rankings should use unrounded values or exact arithmetic where possible. The page should state its tie rule and show enough detail to explain the order. A reader should not have to guess why two equal-looking scores have different ranks.

Sensitivity can be reported honestly without declaring a fragile leader. For example: the top document changes when the policy weight crosses 50. That statement is more informative than showing only one selected scenario. It tells readers which preference drives the result and how close the chosen method is to the boundary.

There are limits to what weights can do. In our table, B has every feature that C has, plus R. With nonnegative weights, C cannot score above B. If R receives zero, they can tie. Giving C a weight of 100 and the other features zero makes B and C tie at 100 while A scores zero.

This is a useful check against the idea that any desired order can always be created by weights alone. Some orders are impossible under a fixed feature table and nonnegative weights. To make C strictly beat B here, someone would have to change the features, allow negative weights, or add another rule. That is a method change worth disclosing.

A sensitivity review should test a reasonable range of priorities, not a collection chosen only to preserve the preferred result. Explain why the tested range fits the research question. A stable result across several scenarios means only that the order survived those scenarios. It does not show that the underlying evidence is accurate or sufficient.

## Hard Requirements Versus Weighted Scores

A weighted score allows one feature to offset another. In our first method, the policy feature alone outweighs the other two together. That is a choice about tradeoffs. Some research questions do not allow that kind of exchange.

Suppose the task requires an explicit policy statement before a record can enter the next step. Use P as a gate: records without the observed statement are outside that particular review. Only A passes within this selected set. Its inclusion does not prove that its sponsor alone has the practice. It shows which selected record meets the stated entry rule.

If the task instead requires individual outcome rows, R becomes the gate and only B passes. There is no need to manufacture a three-place ranking when only one record remains. Describe the result as one qualifying document in the inspected set. Then explain what additional sources would broaden that set.

A gate and a weight solve different problems. A weight expresses a tradeoff. A gate sets a minimum condition. Publishing a method that uses one while describing the other can mislead readers. For instance, a high total score does not prove that every feature met a required standard unless the rules actually require that.

In investment research, some unresolved questions may be too important to average away. A researcher might need the governing agreement before comparing a particular contract term. A strong biography would not replace that missing agreement. This is an example of a research gate, not a statement that every investor must use one universal scoring rule.

Write gates before scoring and explain their purpose. Show excluded records and the reason for exclusion. Do not quietly remove inconvenient entries and then call the remaining list the whole market. An exclusion can be sensible while still limiting what the final result means.

## Missing Data and Sponsor Ranking Limits

Unknown evidence and an observed absence within a defined record require separate treatment. An inaccessible page is unknown. A fully read page that does not contain a specified feature earns zero only under our narrow content test. A missing performance history cannot be assigned a zero return. Those are different kinds of information.

The scope should be clear enough that another reviewer can challenge it. If the task expands from one page to an entire available research packet, the old zeros must be rechecked. They do not automatically carry forward. The object being measured has changed, even if the sponsor name is unchanged.

A hypothetical example shows the effect. Suppose a later record set for A adds outcome rows that meet the same R test. This is an invented change, not a claim about new company material. Under the 60, 20, 20 method, that expanded record set would score 80 instead of 60. B would remain at 40 and C at 20 if their records stayed fixed.

That change would come from evidence scope, while our earlier leader change came from weights. Preserve these causes separately. Otherwise, readers cannot tell whether the ranking moved because a company disclosed something new, a researcher found an old file, or an editor changed the scoring method.

Several important investment subjects are entirely outside our dataset. We have not measured financial capacity, actual service quality, all-in fees, losses, property conditions, or current offering suitability. They are unmeasured here. We do not assign zero to them, estimate them from document length, or imply that the three-feature score covers them.

The Financial Industry Regulatory Authority, or FINRA, addresses investigation of private placements by covered member firms in Notice 23-08. Its discussion supports examining material representations and underlying evidence with critical care. It does not endorse this feature table or its weights. A public-document score cannot substitute for the work that a specific offering requires. [FINRA Notice 23-08](https://www.finra.org/rules-guidance/notices/23-08).

A missing-data rule should be part of the method before results are calculated. A publisher might withhold a score, exclude a factor across all entries, or show separate disclosures. Each choice changes the comparison in a different way. Readers should see what was removed and what remained. A blank cell should never become an unstated assumption.

## How to Read a DST Sponsor Ranking Methodology

Read the method before the first-place name. Begin with the unit of analysis. Does each row represent a sponsor, a legal entity, a fund, an offering, a broker, or a source document? A table that mixes those units can create false comparisons. An excellent source about one offering cannot silently stand for every product from a firm.

Next inspect the candidate set. How were names found, and which names were left out? Was inclusion based on public access, participation, a commercial relationship, or a documented market definition? A publisher can have a reasonable selection rule and still cover only part of the market. The title and conclusion should reflect that boundary.

Look for a clear date for each important input. A research date on the article is not enough. One row may use a recent file while another relies on an older report. That difference matters when the underlying fact can change. It should be visible even when no perfectly matched set is available.

Then inspect the direction of each score. Does a higher number always mean more favorable under the stated rubric? A count of offerings, a fee percentage, and a complaint count have different meanings. Combining their raw values without conversion can make the largest numerical scale dominate the total.

Ask how the publisher handles uncertainty. A fact may be reported by a company, supported by an underlying contract, or independently checked against another record. These evidence states should not blur into one word such as verified. A manual source check confirms that someone inspected a file; it does not settle every claim inside it.

Read the commercial disclosures beside the method. Compensation, selection rights, or paid placement can affect what readers need to know about a list. This article does not investigate any publisher's business model. The practical point is to make those relationships visible so that readers can assess the context of the research.

Finally, check whether a second person could recreate the result. The needed material includes feature definitions, source locations, raw inputs, weights, exclusion rules, and tie handling. A result that cannot be recreated may still contain useful discussion, but its numerical order deserves less reliance until the missing steps are explained.

A strong method also tells readers what the score cannot answer. A ranking of reporting access does not answer which property to buy. A ranking of selected historical outcomes does not establish future returns. Useful research keeps its conclusion close to the evidence that produced it, even when a broader headline would draw more attention.

## Write a Research Decision That Others Can Recheck

A short decision memo can preserve the method without burying the reader. Start with the task in one sentence. For our exercise, that might be choosing a first document for a policy review or locating individual rows for later analysis. The task should be narrow enough that the chosen features make sense.

Then state the source boundary. Name the exact records, their periods, and the inspection date. Say whether linked files are included. List any materials that were requested but not obtained. This turns a vague claim of extensive research into a file that another reader can understand and extend.

Record the feature table before the weighted result. Include a reason for each coding decision and a way to find its source. Preserve unknown states separately. If a reviewer disagrees with one code, the team can discuss that particular judgment without having to reverse-engineer the whole ranking.

Show at least one reasonable alternative method when priorities are debatable. In this article, the policy-first and rows-first cases produce different leaders. The 50-point boundary explains why. A memo can state which method was used for the next research step while acknowledging the alternative rather than concealing it.

Keep changes in a simple log. Separate revised source facts from revised rules and corrected errors. Date the change, explain its effect, and preserve the prior version when possible. A later reader should be able to tell why a result moved. A new publication date alone does not provide that explanation.

End with the next evidence task. It might be obtaining an underlying report, checking outcome definitions, or reviewing a specific offering agreement. The ranking has then served a practical purpose: it has organized further research. It has not been stretched into an investment recommendation that the available evidence cannot support.

## FAQ

### Why Do Lists of Top DST Sponsors Name Different Leaders?

They may use different sources, candidate sets, features, weights, or missing-data rules. Even the same feature table can produce a different order when the research goal changes. Ask what each list measures before comparing its first-place name. A disagreement in rank does not by itself show that either list found different underlying facts.

### Does This Article Rank Capital Square, ExchangeRight, and Inland as Investments?

No. It codes three features in selected public records and orders those records under stated methods. The firm names identify the source authors. The study does not measure investment quality, safety, fees, financial strength, or future returns. Its scores should not be copied into a sponsor recommendation table.

### Are Equal Weights the Fairest Ranking Method?

Not necessarily. Equal weights treat the chosen features alike, but someone still chose those features. Repeating related measures can give one theme extra influence. A fair explanation should discuss both the feature selection and the weights, then show how reasonable alternatives affect the order.

### Can an Unknown Field Be Scored as Zero?

No. An unknown field stays unknown. A zero in this study records an observed absence under a defined test. In our content study, zero means a specified feature was not observed in a fully inspected selected record. It does not mean the sponsor lacks that practice. Failed access, missing financial data, and unmeasured investment risks remain unknown or outside scope.

### What Does a Tie Mean in a DST Sponsor Comparison?

A tie means the chosen method produced equal exact scores. It does not prove equal quality in every respect. Rounded displays can also look tied when exact values differ. Publish the tie rule, retain enough precision to explain the order, and avoid inventing a winner through an undisclosed tiebreaker.

### Is a Stable Ranking More Reliable?

Stability shows that the order survived the particular changes tested. That can help readers understand method sensitivity. It does not prove that the source claims are true, the candidate set is complete, or the features measure investment quality. Evidence review and method testing address separate questions.

### What Should I Request Before Relying on a Published Ranking?

Request the source list, data dates, feature definitions, weights, treatment of unknowns, exclusions, and tie rules. Ask what the score is intended to measure and which decisions it cannot support. If the ranking leads you toward a particular offering, review that offering's own documents and seek qualified advice for your circumstances.

## About Sequoia Research

Sequoia Research, LLC provides independent research on DSTs and other real estate investment offerings. This educational library explains how to inspect sources, compare investment terms, and separate verified facts from open questions. This article is general research education and does not provide an individual investment, legal, or tax recommendation.

## Sources

- [Capital Square Real Estate Investments](https://capitalsq.com/real-estate-investment/), undated company page; selected record reviewed October 17, 2026. Actual audit reports not inspected.
- [ExchangeRight Investment Track Record](https://www.exchangeright.com/resources/investment-track-record.pdf), August 31, 2026, eight pages; selected record reviewed October 17, 2026.
- [Inland 2025 Private Investments Portfolio Overview](https://inland-investments.com/insights/private-investments-portfolio-overview), landing page covering programs through December 31, 2025; reviewed October 17, 2026. Linked full report not obtained.
- [FINRA Regulatory Notice 23-08](https://www.finra.org/rules-guidance/notices/23-08), May 9, 2023; relevant private-placement investigation discussion reviewed October 17, 2026.


## Related Reading and Tools

- [Largest DST Sponsors (2026): What Size Actually Measures](https://sequoia-research.com/writing/largest-dst-sponsors-2026-what-size-actually-measures)
- [How to Choose DST Sponsor: A Step-by-Step Research Checklist](https://sequoia-research.com/writing/how-to-choose-dst-sponsor-a-step-by-step-research-checklist)
- [Top DST Sponsors for 1031 Exchanges (2026): How to Compare Them Before Your Identification Deadline](https://sequoia-research.com/writing/top-dst-sponsors-for-1031-exchanges-2026-how-to-compare-them-before-your-identification-deadline)