Zenvestly

Methodology

How we source, verify, and update the data behind every comparison on Zenvestly. This is a living document — when our process changes, we update this page.

1. Data Sources

We use only sources that can be re-verified by anyone reading our pages. We do not rely on private datasets, gated reports, or anonymous tips for any number we publish.

  • Vendor pricing pages — primary source for any subscription cost, tier feature, or usage limit. Cited inline with the URL date stamp at time of verification.
  • FRED (Federal Reserve Economic Data) — interest rates, treasury yields, inflation indices. Pulled live by our backend and embedded with the FRED series ID.
  • U.S. Treasury Direct, Frankfurter, official APY/APR disclosures — for yield, currency, and savings-rate data.
  • Public SEC filings, S-1 and 10-K disclosures — when a vendor cost structure or revenue model is referenced beyond their pricing page.
  • Vendor documentation, API references, official changelogs — for feature claims, rate limits, and product capability assertions.

Sources we explicitly do not use as the sole basis for a number: vendor marketing landing pages (use the pricing page), social media posts, third-party aggregator sites that don't themselves cite a primary source.

2. Comparison Framework

Every head-to-head review on Zenvestly applies the same structural framework. This is what makes our comparisons reproducible across categories — the framework doesn't change between an AI tool review and an HYSA comparison, only the inputs do.

  • Headline price + true cost. The number on the pricing page, plus the realistic per-seat or per-use cost once base plans, add-ons, and usage caps are accounted for. Often the “real” number differs from the headline — we surface that.
  • Feature scope matrix. A side-by-side row for every meaningful feature axis. Where one product covers a feature the other does not, the gap is called out, not glossed.
  • Free-tier reality. What you can actually do without paying, with specific limits (messages per day, file size, seats, API calls).
  • Weakness section. Mandatory. Every review must articulate a real downside per product. If we cannot, we have not tested or researched the product enough to publish.
  • Verdict by use case. Not “Product A wins.” We assign specific use cases to specific products with the trade-off named. A reader should know which product is right for which situation, not which product is universally best.

3. Verification & Update Cadence

Pricing and rate data ages. We use a date-stamped verification approach so a reader can see how recently any specific number was checked.

  • Rate-driven pages (HYSA comparison, fee grids): re-pulled and re-verified weekly against the source APIs. The most recent verification date appears on the page.
  • Pricing-driven reviews (AI tool, SaaS comparisons): the pricing section carries a “verified [month] [year]” tag. We re-verify on a rolling 60-day cadence and update the date stamp on the post when checked.
  • Feature claims: re-verified at the same cadence as pricing for the same product. When a feature is added or removed by the vendor, we update the comparison matrix and add an updated date.
  • Reader corrections: any factual correction submitted via email is investigated within one week. If the correction is valid, the page is updated with a corrections note at the bottom.

4. What “Verified” Means

When a Zenvestly page says a number was verified on a date, that means a human editor or an automated checker re-pulled the source on that date and the number matched what is on the page. It does not mean the number is guaranteed accurate forever — vendors change pricing without notice, and rates move daily. The verification date is a freshness signal, not a guarantee.

5. Computer-Assisted Drafting

Zenvestly uses modern language-model and analysis tooling for drafting and data assembly. The framework above — what we compare, what we cite, what we refuse to publish, how we update — is human-authored and applied uniformly. We do not consider tooling choice a substitute for editorial standards. Every number is traced back to the source list above before publication. Pieces that fail the quality gate (length, matrix count, weakness section, source citations) are held back regardless of how they were drafted.

6. Limits We Acknowledge

  • We cannot test every product hands-on with every workflow. When our comparison relies on documented features rather than direct testing, we say so on the page.
  • Rate snapshots become stale within hours for the most volatile assets. For anything time-sensitive, the page links to the live source so a reader can re-check.
  • Our scope is intentionally narrow (AI tools, SaaS, fintech data). We do not cover stock picks, individual security analysis, retirement planning, or tax strategy. Those areas require licensed expertise we do not claim.

Contact

Methodology questions, source disputes, or process feedback: [email protected].

Last updated: 2026-05-03