How to Use Andi for Source-preferring comparisons

Andi works well for source-preferring comparisons when you run it like production work: locked brief, SOURCE facts, then review before publish. Andi is an AI search assistant that answers questions from the web with a conversational interface and no ads or tracking. Public pricing plans are not verified. Treat answers as research aids and verify critical facts. Start at /explore/andi.

This guide focuses on source-preferring comparisons in detail. Related Andi articles: /blog/how-to-use-andi-for-follow-up-narrowing-queries, /blog/how-to-use-andi-for-how-to-step-extractions, /blog/how-to-use-andi-for-definition-and-glossary-briefs.

When this workflow is the right job

Use source-preferring comparisons when the deliverable is specifically this Andi job. Switch to standards and compliance lookups when that workflow already owns the asset.

Step by step workflow

1. Brief Source-preferring comparisons

Write what must stay true for source-preferring comparisons in Andi before settings or spend.

Brief: Source-preferring comparisons
Keep: verified SOURCE facts only
Avoid: invented pricing or features
Success: one reviewable output

2. Open Andi for Source-preferring comparisons

Use the Andi surface that owns source-preferring comparisons. Do not mix a neighboring workflow in the same pass.

Surface: Source-preferring comparisons
Start: pilot with one representative input
Plans: andisearch.com

3. Pilot Source-preferring comparisons

Run a single source-preferring comparisons pilot. Score clarity, grounding, and whether the output is reviewable.

Pilot: Source-preferring comparisons
[ ] SOURCE facts match
[ ] Output reviewable
[ ] Settings logged

4. Refine Source-preferring comparisons

Change one source-preferring comparisons dimension only. Save a template from the best run.

Refine: Source-preferring comparisons
Change: one control only
Keep: SOURCE and success criteria

Practical source-preferring comparisons examples

WCAG 2.2 bullets

Scenario:
A researcher uses Andi for Source-preferring comparisons about "WCAG 2.2 bullets".

Objective:
Produce a sourced brief for WCAG 2.2 bullets with verification flags.

Inputs:
- Query/ticker/topic for WCAG 2.2 bullets
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for WCAG 2.2 bullets → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for WCAG 2.2 bullets with bullets, caveats, and items needing primary-source checks.

W3C/WAI only

Scenario:
A researcher uses Andi for Source-preferring comparisons about "W3C/WAI only".

Objective:
Produce a sourced brief for W3C/WAI only with verification flags.

Inputs:
- Query/ticker/topic for W3C/WAI only
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for W3C/WAI only → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for W3C/WAI only with bullets, caveats, and items needing primary-source checks.

Time range 12m

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Time range 12m".

Objective:
Produce a sourced brief for Time range 12m with verification flags.

Inputs:
- Query/ticker/topic for Time range 12m
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Time range 12m → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Time range 12m with bullets, caveats, and items needing primary-source checks.

How-to five steps

Scenario:
A researcher uses Andi for Source-preferring comparisons about "How-to five steps".

Objective:
Produce a sourced brief for How-to five steps with verification flags.

Inputs:
- Query/ticker/topic for How-to five steps
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for How-to five steps → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for How-to five steps with bullets, caveats, and items needing primary-source checks.

Glossary brief

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Glossary brief".

Objective:
Produce a sourced brief for Glossary brief with verification flags.

Inputs:
- Query/ticker/topic for Glossary brief
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Glossary brief → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Glossary brief with bullets, caveats, and items needing primary-source checks.

Fact-check list

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Fact-check list".

Objective:
Produce a sourced brief for Fact-check list with verification flags.

Inputs:
- Query/ticker/topic for Fact-check list
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Fact-check list → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Fact-check list with bullets, caveats, and items needing primary-source checks.

Follow-up narrow

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Follow-up narrow".

Objective:
Produce a sourced brief for Follow-up narrow with verification flags.

Inputs:
- Query/ticker/topic for Follow-up narrow
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Follow-up narrow → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Follow-up narrow with bullets, caveats, and items needing primary-source checks.

Domain prefer

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Domain prefer".

Objective:
Produce a sourced brief for Domain prefer with verification flags.

Inputs:
- Query/ticker/topic for Domain prefer
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Domain prefer → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Domain prefer with bullets, caveats, and items needing primary-source checks.

Uncertainty label

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Uncertainty label".

Objective:
Produce a sourced brief for Uncertainty label with verification flags.

Inputs:
- Query/ticker/topic for Uncertainty label
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Uncertainty label → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Uncertainty label with bullets, caveats, and items needing primary-source checks.

Citation URLs

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Citation URLs".

Objective:
Produce a sourced brief for Citation URLs with verification flags.

Inputs:
- Query/ticker/topic for Citation URLs
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Citation URLs → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Citation URLs with bullets, caveats, and items needing primary-source checks.

Executive 5 bullets

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Executive 5 bullets".

Objective:
Produce a sourced brief for Executive 5 bullets with verification flags.

Inputs:
- Query/ticker/topic for Executive 5 bullets
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Executive 5 bullets → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Executive 5 bullets with bullets, caveats, and items needing primary-source checks.

Developer notes

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Developer notes".

Objective:
Produce a sourced brief for Developer notes with verification flags.

Inputs:
- Query/ticker/topic for Developer notes
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Developer notes → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Developer notes with bullets, caveats, and items needing primary-source checks.

Caveat banner

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Caveat banner".

Objective:
Produce a sourced brief for Caveat banner with verification flags.

Inputs:
- Query/ticker/topic for Caveat banner
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Caveat banner → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Caveat banner with bullets, caveats, and items needing primary-source checks.

Ticker ALT signals

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Ticker ALT signals".

Objective:
Produce a sourced brief for Ticker ALT signals with verification flags.

Inputs:
- Query/ticker/topic for Ticker ALT signals
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Ticker ALT signals → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Ticker ALT signals with bullets, caveats, and items needing primary-source checks.

Web traffic MoM

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Web traffic MoM".

Objective:
Produce a sourced brief for Web traffic MoM with verification flags.

Inputs:
- Query/ticker/topic for Web traffic MoM
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Web traffic MoM → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Web traffic MoM with bullets, caveats, and items needing primary-source checks.

Hiring surge flag

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Hiring surge flag".

Objective:
Produce a sourced brief for Hiring surge flag with verification flags.

Inputs:
- Query/ticker/topic for Hiring surge flag
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Hiring surge flag → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Hiring surge flag with bullets, caveats, and items needing primary-source checks.

Social spike

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Social spike".

Objective:
Produce a sourced brief for Social spike with verification flags.

Inputs:
- Query/ticker/topic for Social spike
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Social spike → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Social spike with bullets, caveats, and items needing primary-source checks.

Starter vs Pro

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Starter vs Pro".

Objective:
Produce a sourced brief for Starter vs Pro with verification flags.

Inputs:
- Query/ticker/topic for Starter vs Pro
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Starter vs Pro → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Starter vs Pro with bullets, caveats, and items needing primary-source checks.

Monthly digest

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Monthly digest".

Objective:
Produce a sourced brief for Monthly digest with verification flags.

Inputs:
- Query/ticker/topic for Monthly digest
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Monthly digest → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Monthly digest with bullets, caveats, and items needing primary-source checks.

Primary-source flag

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Primary-source flag".

Objective:
Produce a sourced brief for Primary-source flag with verification flags.

Inputs:
- Query/ticker/topic for Primary-source flag
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Primary-source flag → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Primary-source flag with bullets, caveats, and items needing primary-source checks.

Competitor compare

Scenario:
A researcher uses Andi for Source-preferring comparisons about "Competitor compare".

Objective:
Produce a sourced brief for Competitor compare with verification flags.

Inputs:
- Query/ticker/topic for Competitor compare
- Preferred sources/domains
- Output format
- Time scope

Workflow:
Ask focused question → Review answer for Competitor compare → Flag unverified claims → Follow up to narrow → Save brief

Requirements:
- Stay within verified Andi capabilities; do not invent features.
- Confirm live plan notes on andisearch.com before promising volume.
- Change one variable between iterations.
- Human-review before external publish, send, billing, or clinical/legal use.
- Not investment advice when market data is involved.

Expected output:
A brief for Competitor compare with bullets, caveats, and items needing primary-source checks.

How to improve source-preferring comparisons

Cut noise from source-preferring comparisons by removing extra adjectives while preserving SOURCE facts in Andi.

Raise quality by insisting on a single success check before debating style.

Make review easier by labeling fields that must never change.

Speed iteration by cloning the last good run and altering only one control.

Stabilize outputs by pinning settings after the pilot is approved.

Reduce rework by rejecting drafts that invent claims.

Improve handoffs by recording which control produced the best result.

Harden the workflow by testing an incomplete input before trusting defaults.

Prompting and usage guidance

Name the source-preferring comparisons job, audience, and success check before opening Andi.

Paste only verified facts under SOURCE so Andi cannot invent details.

Specify the deliverable shape up front.

Call out fixed details versus flexible style choices.

Ask Andi to flag unsupported claims before you accept the draft.

Limitations to respect

Check Andi plan gates for source-preferring comparisons on andisearch.com before you promise timelines.

Keep drafts unpublished until a human confirms SOURCE facts.

Andi can be wrong. Treat source-preferring comparisons as provisional until review.

If documentation is silent on a claim, leave it out rather than guessing.

Practical tips for this workflow

Pilot once before batching source-preferring comparisons in Andi.

Keep a reusable template with variables for source-preferring comparisons.

Separate creative instructions from SOURCE facts.

Log settings from the best run.

Common mistakes

• Skipping the pilot run before scaling volume
• Inventing pricing, quotas, or features not on official pages
• Mixing unrelated workflows in one session
• Publishing without a human review gate

Treat source-preferring comparisons in Andi as a production workflow: brief, pilot, refine, then ship with review. Related reading: /blog/how-to-use-andi-for-follow-up-narrowing-queries, /blog/how-to-use-andi-for-how-to-step-extractions, /blog/how-to-use-andi-for-definition-and-glossary-briefs.