From your data source — surveys you've run, third-party APIs like Levels.fyi or Glassdoor with appropriate licensing, or your own database of candidate placements. SleekRank only renders what you provide; sourcing, licensing and verifying numbers is your team's job. Always disclose the source in the methodology block — readers and search engines both reward that transparency.

Either store converted values in additional columns (median_usd, median_eur) and render them per row, or compute conversions in your data pipeline before SleekRank reads the source. Live FX conversion at render time is possible via an API call but adds latency and an external dependency. Pre-computing and storing the values in the source is more reliable for content that doesn't need second-by-second accuracy.

Add Occupation JSON-LD if the page is a salary reference (typical case). Use JobPosting only if the page is an actual job ad. Inject role, region, salary range and currency via selector or meta mappings. The Rich Results Test will catch malformed schema; validate after the first refresh and again after any template change. Wrong schema type can suppress salary-related rich results.

Put the methodology on the base page once — sample collection, time period, sample size, any anonymization, exclusions. Every generated salary page inherits it automatically with identical wording. That solves the consistency problem hand-built salary posts struggle with: different posts, different disclosure depths, different language, which becomes a credibility risk for any publisher serious about salary content.

Compute the YoY delta in your data pipeline and store it in a column — yoy_pct or yoy_delta. SleekRank renders whatever fields you provide. Visualize it as an arrow with a percentage on the page, surfaced near the median figure. Readers value YoY context because it tells them whether the snapshot is rising, flat or falling, which raw numbers alone don't communicate.

No. SleekRank reads from your data source. Sourcing, surveying and verifying is entirely your responsibility. Pair with a small validator that flags rows where last_reviewed is more than twelve months old, where sample_size is below a threshold (sub-50 sample sizes lose statistical meaning), or where high is more than 4x median (likely a data entry error). SleekRank renders rows; quality control is yours.

Add seniority as another dimension — software-engineer-senior-us, software-engineer-staff-us. The matrix grows but stays clean. For long-tail outliers, store p90 or p95 values in additional columns and surface them on the page with the right framing. Don't bury senior-IC data in a single role-region cell; it warps the median for the broader role and confuses readers.

Yes, by editing the base page. Because methodology lives once and every salary page inherits it, swapping a long disclosure for a shorter one and measuring impact on bounce rate is a single edit. Use a feature flag or a simple URL parameter for the test. Hand-built sites can't run that test cleanly because the disclosure varies across posts.