Sample KONSİL report

A real run, frozen on 20 September 2026. The images below are taken from the app's own screen; every number and text is that run's output.

The report is published with its owner's permission. Candidate researchers' names, institutions, ORCID iDs and contact details are masked on this page; only verified KONSİL users see them. Source publications are open records.

The verdict and evidence section comes from the 20 September measurement run; the design and team sections come from an earlier run of the same idea. Both are frozen. The verdict run was in Turkish: its report prose in the images (findings, gap, recommendations) is translated into English by hand; verbatim quotes and paper titles are the papers' own English. The design and team sections come from an English run.

1. Idea entered

“In a rat model of doxorubicin-induced cardiotoxicity, late (rescue/therapeutic) dapagliflozin initiated only AFTER cardiac dysfunction is echocardiographically established: its capacity to reverse functional decline and halt fibrosis progression, compared head-to-head with a concurrent-preventive protocol, with blinded outcome assessment — an experimental animal study”

2. Verdict and claims

Verdict and claims

The idea is first split into claims. The verdict is one sentence, and the reason for it sits right below.

The verdict block in the app: Partly done; four claims and their status; the gap identified; what the verdict rests on.

1
Verdict. One sentence: “Partly done.” Not to approve, to test: it says how much of the idea has already been done by others.
2
Breakdown. The verdict comes from the claims' status: of 4 claims, 1 already studied, 2 partly known, 1 still open.
3
Claim ledger. The idea is split into separate claims and each is tested on its own. Status is given by shape and text, not by colour alone.
4
The gap identified. Exactly what has not been done: here, a blinded comparison of late-started (rescue) treatment against the preventive protocol.
5
Basis. 20 publications were evaluated in detail; 8 read in full text, 12 from the abstract; 6 verbatim quotes were verified in the article text.

3. A closer look at one claim

A closer look at one claim

Each claim card carries its own evidence: status, finding, the closest paper and the evidence itself.

One claim card in the app: status badge, finding, closest evidence and the evidence button.

1
Status. Each claim's status sits in the card's corner: Already studied, Partly known, Still open or Undetermined.
2
Finding. Every source named in the finding links to the paper itself; there is no room for an invented reference.
3
Closest evidence. The paper closest to the claim, by title and year.
4
Role label. The paper's relation to this claim is labelled. Here, “Design precedent”: the method is similar, the question is not the same.
5
Evidence. Opens the verified quotes the claim rests on.

4. What carries the verdict?

What carries the verdict?

Evaluated publications are grouped by role; the ones that carry the verdict stand first.

Evaluated publications in the app: role labels, source, year and each paper's rationale.

1
Scope. How many publications were evaluated in detail, and that each one's role for the idea is stated, is said up front.
2
Table. The same publications also open side by side in a comparison table.
3
Groups. Publications are grouped by role; “Carry the verdict” comes first, with the quotes they rest on.
4
Role and text. Each entry shows its role, year, journal and whether it was read in full text or from the abstract.
5
Rationale. One sentence says why the paper has this role, and which dimensions differ (here Task and Outcome).

5. Conclusions and recommendations

Conclusions and recommendations

The gap found, the protocol built around it and what not to rely on, side by side.

1
Research gap. The same sentences as in the verdict's rationale: what has not been done, with sources.
2
Protocol suggestions. How the study should be built around the gap: comparison, blinded measurement, outcomes.
3
Caveats. The limits of the findings are written plainly: which studies used a preventive protocol, which details are missing.

6. Proposed study design

Proposed study design

The design is proposed from the evidence; what each field is for is written next to it.

1
Design summary. The proposed design in one paragraph, with the status of the novelty question. Here the earlier run could not settle it (evidence insufficient), so the design is framed as a feasibility/pilot study.
2
Skeleton. Next to each field, what it is for: here, the skeleton of the study.
3
Rationale. Why this design and not another.
4
Sequence. From the start to the analysis, step by step. Decisions not yet made are not invented; they are left to you.

7. Team suggestions

Team suggestions

Candidates come by role, ranked by ability to collaborate, not by prestige.

1
Institution. The institution comes from the person's own ORCID record or the address line on their papers; users are told to verify it before writing.
2
Candidate. The candidate's name is hidden on this page; in the app only verified members see it.
3
Fit. Field match, reachability and shared language are shown separately; the execution-fit score sits in the ring beside them.

Not in this example yet

Evidence-based power analysis (the expected value in a sample-size calculation coming from similar studies, with their own sentences) was added after this frozen example; the example's power section is in the earlier format, so it is not shown here. How it works is in the “What about n?” film.

Try it with your own idea

The first 20 research ideas selected in the closed beta will be evaluated free of charge.

Apply to the closed beta