Trust and Transparency
The Swoopr Research Loop
Direct answer: The Swoopr Research Loop is the eight-stage process the Swoopr Editorial Team uses when building educational research content: Define, Understand, Measure, Compare, Value, Stress-Test, Document, Review. It is a repeatable way to organize research writing, not an investment strategy, not a stock-picking system, and not a promise about future performance.
Most of the guides, research frameworks, and tools on Swoopr Investment are built the same way underneath, even when the topic is a single valuation ratio or a full comparison framework. The Swoopr Research Loop names that underlying process so readers can see the actual steps behind a piece of content, not just the finished page.
What the Research Loop is not
The Research Loop describes how we research and write, not a system for picking winners. It does not rank securities, does not generate buy or sell signals, and is never used to justify a Swoopr-branded index, portfolio, or track record. See Financial Disclaimer and Evidence & Validation for why we publish auditable process evidence instead of performance claims.
1. Define
State the actual question a page or tool answers, in plain terms, before writing anything. "How does a Sharpe ratio work" and "how do I compare two ETFs on cost" are different questions with different right answers. A vague question produces a vague, unfocused page. See Where to Start for how we scope a topic against what a reader is actually trying to decide.
2. Understand
Establish the mechanics before the numbers: what the concept is, how it's defined, and how it actually behaves in practice, including edge cases and common misreadings. This stage draws on primary sources (regulators, exchanges, standards bodies, official documentation) per our Citation & Sources Policy, not secondary paraphrase.
3. Measure
Turn the concept into something that can actually be calculated: the formula, its inputs, and its units. Any calculator or interactive tool traces back to a formula stated in plain text on the page, implemented in a single, dedicated, unit-tested code module, never duplicated logic that can silently drift out of sync, see Evidence & Validation for how that's enforced.
4. Compare
Put the measurement next to the alternatives: other formulas that answer a related question, other assets or instruments a reader might be evaluating, or other methodologies that reach a different number from different assumptions. This is where a page distinguishes what's actually being measured from what a reader might assume it means.
5. Value
Work through what the number or comparison implies, using a concrete, newly constructed worked example with invented neutral figures, not a repeated stock pick or a copied example from anywhere else. A value judgment about a formula's output is always shown as conditional on stated assumptions, never as a standalone conclusion.
6. Stress-Test
Ask where the method breaks: unusual market conditions, edge-case inputs, small-sample effects, or assumptions that stop holding. Every calculator's assumptions and limitations are disclosed on its own page, and every backtested or simulated result is labeled as such, see Risk Disclosure and Data & Methodology for what simulated results do and don't represent.
7. Document
Write down the sources, the formula, the assumptions, and the worked example on the page itself, in a References section where the page makes a claim that depends on an external authority, per our Citation & Sources Policy. A reader should be able to verify a claim without taking Swoopr's word for it.
8. Review
Revisit published content on a cadence, faster for anything tied to a tax rule, regulation, or protocol specification that can change, per our Editorial Policy. A page's "Updated" date reflects an actual material revision, and factual errors are corrected under the Corrections Policy, with the correction logged on Evidence & Validation.
Methodology version
This is version 1.0 of the published Swoopr Research Loop, first published August 21, 2026. If the stages or their descriptions change materially, that revision is recorded in the methodology version history on Evidence & Validation.
Frequently Asked Questions
Is the loop applied to every page on the site?
The stages describe how content is researched and written, and how completely each applies depends on what the page is. A calculator exercises the measure and stress-test stages heavily because it has a formula and disclosed limitations. A glossary definition exercises define and understand, and carries a review cadence rather than a worked example. A page that skips a stage because it does not apply is different from one that skipped it because the work was not done.
How does this differ from the research process published for readers?
They share the same eight stages applied to different objects. This one governs how a page or tool gets built: the question it answers, the sources behind it, the formula it implements and the review it receives. The reader-facing version applies the same sequence to researching a company or an asset. Using one framework for both is deliberate, since the discipline being described, stating the question before gathering evidence, is the same discipline.
What happens when a stage cannot be completed?
The gap is recorded rather than filled in. The clearest example on the site is the research notes library, where primary-source data was not available in the build environment, so the notes were published as briefs stating exactly which documents the finished analysis will draw from, with no figures asserted. Publishing an incomplete page that says so is treated as preferable to completing it with material that was not verified.
Why does the value stage use invented figures?
Because a worked example exists to demonstrate a mechanism, and a real security drags in facts that will be stale within months and implications the page is not making. Neutral invented figures let the arithmetic be followed exactly while making clear that nothing is being asserted about any actual company. Every such example on the site is labeled as hypothetical for that reason, and the labeling is part of the convention rather than a disclaimer added afterward.
What counts as a material revision that changes a page's updated date?
A change to what a reader actually receives: revised prose, corrected figures, a changed formula, new sections. Site-wide operations that rewrite files without changing content, such as a navigation sync or an index regeneration, do not qualify, and the date stamping is driven by a fingerprint of the page's main content rather than by whether the file changed. That distinction is enforced automatically, so a date cannot drift without a corresponding content change.
How is the loop enforced rather than just described?
Through automated checks on the artifacts each stage produces. Calculator logic must live in a separately tested module; pages making externally verifiable claims must carry a References section under a fixed heading; structured data must match the visible content word for word; updated dates must correspond to a real content change. None of these checks judges quality. They confirm the artifact exists and is consistent, which is what makes an absence visible.
Does the loop apply to interactive tools as well as written guides?
It applies more strictly, because a tool produces numbers a reader may act on. The measure stage requires the formula to be stated in plain text on the page and implemented once in a tested module. The stress-test stage requires the assumptions and limitations to be disclosed on the tool's own page, and any simulated or backtested output to be labeled as such. A calculator without a disclosed limitations section has skipped a stage.
What sends a page back to an earlier stage?
A change in what the page depends on. A regulation, tax treatment or protocol specification that changes sends the page back to understand and measure; a confirmed factual error sends it back through the same route under the Corrections Policy. A new competing page covering the same topic sends both back to define, since two pages disagreeing on the same fact is a problem to resolve rather than a difference of emphasis.
Why is the methodology versioned rather than edited in place?
So a change to the process itself is visible instead of silent. A page produced under one version of the process and a page produced under a later one were built to different standards, and quietly rewriting the description would erase that. Versioning also makes the change reviewable: a material revision to the stages or their descriptions is recorded in the version history rather than being discoverable only by having read the previous text.