Use caseMy sources disagree

Find out exactly where your sources stop agreeing

Four accounts of the same subject, three different stories. Get the agreement, the conflict and the missing evidence separated — with both sides kept.

Free plan · No card required

The subjects are the sources

  • source 1Vendor migration page
  • source 2Independent review
  • source 3Practitioner forum thread
  • source 4Benchmark report

Where sources agree

All four treat migration effort, not licensing, as the real switching cost source 1 source 2 source 3 source 4

Where sources differ · both kept

  • source 1Vendor migration page

    “Most teams are live in under three days”

  • source 3Practitioner forum thread

    Two to five weeks, across six accounts

Still open

Neither states the team size it describes, so the two figures may not be comparable at all.

The situation

The disagreement isn’t noise. It’s the thing you need to understand.

A vendor page, an independent review and a practitioner thread describe the same subject and give you incompatible pictures. Averaging them loses the information; picking the one you like is how you get surprised later.

  • Typical shape

    Vendor claim vs independent review

    Both describe the same capability. One is selling it — neither is useless, and only one is advocacy.
  • Typical shape

    Two reports, different numbers

    Often because they measured different things. The useful output is seeing the figures were never comparable.
  • Typical shape

    Official docs vs practitioner accounts

    Documentation describes the intended path; threads describe what happened. The gap between them is the finding.
  • Typical shape

    An AI answer vs the sources

    A captured ChatGPT or Claude conversation can sit in the set as AI-generated context. It never counts as one of the independent voices.

What this job does, and doesn’t

It won’t settle the argument for you

Nothing decides which source is right. The brief keeps both readings with their citations, because that is what the evidence supports.

It makes the disagreement legible

Agreement, conflict and missing evidence are separated and named — so you can see whether the argument is about evidence, about definitions, or about nothing at all.

Start here

Point the comparison at the sources

Comparison covers both kinds of comparison, and your wording decides which one you get. Ask about the sources and SourceFrame compares what they say; name options and it compares the options.

Web app · Comparison

Where do vendor claims and independent reviews disagree on how long a Jira to Linear migration takes?

Find sources & createMode · Comparison

The link fills in the prompt only — choose Comparison in the composer before you create.

Route in
  1. Wording about the sources
  2. Comparison
  3. Source Comparison Brief

Discovery can find the conflicting sources for you. If you already have the pages, capture them and ask the same way. Two or more usable sources are needed.

What decides it

In Comparison, the wording decides:

Comparison · “Where do these pages agree and disagree about migration effort?”
Source Comparison Brief
Comparison · “What do the Linear and Jira reviews say about migration effort?”
Source Comparison Brief
Comparison · “Compare Linear and Jira”
Comparison Brief

Naming products doesn’t turn it into an option comparison: while the wording points at the sources, the sources stay the subject.

A different comparison

Sources as the subjects, not options

Source Comparison Brief asks

“What do these sources agree and disagree about?”

Versus

Comparison Brief asks

“How do these options compare on the criteria that matter?”

What you get

Both readings, kept and attributed

Three cuts from one Source Comparison Brief. What’s inside a Source Comparison Brief

  • Where they line up

    Where sources agree

    Migration effort dominates the switching decision

    All four sources treat data and workflow migration, not licence cost, as the real barrier source 1 source 2 source 3 source 4.

    What the sources converge on, cited to each of them.

  • Where they split

    Where sources differ

    Disputed:

    How long migration actually takes

    The vendor page says most teams are live in under three days, with no sample size given source 1; six practitioners in the forum thread report two to five weeks, mostly spent on historical tickets source 3.

    A difference keeps both readings with their citations — the brief doesn’t pick one.

  • What none of them settles

    Open questions

    • Do the vendor and the thread describe teams of comparable size? Neither states it.

    Listed as an open question, not smoothed over.

How SourceFrame handles the evidence

What kind of source it is becomes part of the reading

Divergence is preserved

Conflicting readings sit side by side with their citations. Nothing is averaged into a claim no source made.

Source quality, by dimension

Each source shows its perspective and, separately, its evidence depth. Advocacy is a perspective and assertion-only is an evidence depth — neither is a score.

AI chats aren’t a voice

A captured conversation is cited as AI-generated context, so it can’t manufacture agreement between independent sources.

One voice isn’t consensus

Where only one source speaks to a point, that goes in Evidence gaps & caveats rather than being read as everyone agreeing.

A worked run

Four accounts of one migration

  1. The set and the focus

    • source 1

      Vendor migration page

      Perspective · advocacy
    • source 2

      Independent review

      Perspective · balancedEvidence · example-based
    • source 3

      Practitioner forum thread

      Evidence · assertion-only
    • source 4

      Benchmark report

      Evidence · data-heavy

    Comparison · wording about the sources

    Where do these pages agree and disagree about migration effort?

    Captured or discovered, the four sources are the subjects.

  2. What came back

    • Agreement: Agreement on what the cost is — all four say migration, not licensing source 1 source 2 source 3 source 4
    • Difference: An order-of-magnitude disagreement on how long it takes source 1 source 3
    • Difference: A second difference: whether automation rules carry over or are rebuilt by hand source 2 source 3
    • Open: Neither timeline states its team size, so the figures may never have been comparable

    Both timelines stay in the brief with their citations. It doesn’t pick one.

  3. The next meaningful action

    1. Add sources
    2. variant 2

    Add one measured case study with a stated team size, and read the updated variant beside the original, which stays Primary. If you have to act despite the disagreement, Turn into… a Recommendation Brief, which states its status accordingly.

Where you can go next

  • Primary brief type

    Source Comparison Brief

    The full structure, and how comparing sources differs from comparing options.

  • Different comparison

    Comparison Brief

    When the options are the subject and you need shared criteria instead of perspectives.

  • Closely related job

    Check an AI answer

    The same mechanics with a captured AI conversation in the set, as context beside the independent sources.

  • If themes matter more

    Summarize a source set

    Same kind of reading, reported as themes and tensions rather than point-by-point agreement.

Stop averaging sources that disagree.

Free plan · No card required