Migrate from Scispace to Elicit.
2 documentation-derived translation patterns — what carries over and what to watch for. Cited to the Feature Parity Map; the audit tells you whether the move is worth it.
This is the same distinctive workflow on both tools: lay papers out as rows in a table and add custom columns whose natural-language instructions pull a specific data point (methods, sample size, results, limitations, key quotes) across every paper. A team running structured literature reviews in SciSpace can cut it and rebuild the matrix in Elicit. Pull the paper set into Elicit (search natively, or bulk-import your SciSpace shortlist via RIS/BIB), open the table view, and click Column to add custom columns — each takes a name plus multi-paragraph instructions, supports open-ended, yes/no/maybe, and multiple-choice types, and can be saved as a reusable Preset; Elicit extracts across up to 1,000 papers in one pass and attaches a supporting quote from the source to every cell. Net result: the SciSpace 'add a custom column, extract across all papers, compare side by side' loop is reproduced in Elicit, with per-cell source quotes for auditability. Keep Elicit; cut SciSpace.
- Warning: Column budgets differ and Elicit's are tighter per pass: SciSpace allows up to 50 columns (including default suggestions), while Elicit gates columns-at-a-time by plan (Basic 2, Pro 20, Scale 30, Enterprise 40). A very wide SciSpace review — and especially SciSpace's Deep Review meta-analysis mode (reported up to ~100 columns) — will not map one-to-one onto a single Elicit table; split it into multiple extraction passes or confirm a high-enough Elicit tier before cutover.
- Warning: Migrate the papers, not the analysis: there is no table-to-table importer. Export the SciSpace comparison table/collection to CSV/Excel/RIS, import the references (RIS/BIB) into Elicit, then re-author the columns — the extracted-cell contents do not carry over and Elicit will re-run extraction.
- Warning: Extraction sub-limits live in Elicit's help center, not as a headline monthly quota: e.g. Pro caps uploaded-PDF extractions at roughly 1,200/year. For PDF-heavy reviews, check the current per-plan extraction allowance so you are not capped mid-project after dropping SciSpace.
- Warning: Source breadth: Elicit uses a paper's full text when a PDF is available and the abstract otherwise; if SciSpace was extracting from full-text PDFs you uploaded, make sure those PDFs are attached in Elicit so column answers are drawn from full text rather than abstract-only.
Both tools start research the same way: type a plain-language question and get back papers ranked by meaning, not just keyword overlap. A team paying for SciSpace just for this retrieval step can cut it and run the same first move in Elicit. Open Elicit, use the paper search (semantic or keyword mode) over its 138M+ index sourced from Semantic Scholar, OpenAlex, and PubMed; each result carries a query-specific AI summary and filters for year, study type, journal quartile, and PDF availability, so the 'find relevant studies' job that SciSpace's literature search did is covered. The practical upgrade on migration: Elicit's search is free and unlimited across all plans (credits were retired in the August 2025 overhaul), whereas SciSpace's free Basic tier caps literature searches at roughly 5 — so the day-to-day searching that pushed users onto a paid SciSpace plan no longer needs a paid seat. Keep Elicit; cut SciSpace.
- Warning: Corpus size and composition differ: SciSpace indexes 280M+ papers, Elicit 138M+ (deduplicated from Semantic Scholar / OpenAlex / PubMed). A handful of papers that surfaced in SciSpace — especially non-deduplicated or niche records — may not appear in Elicit, so re-run any saved/critical SciSpace queries in Elicit and confirm key references are present before retiring the SciSpace account.
- Warning: SciSpace search has no native 'saved query' export — there is no project file to migrate. Carry over your search strings/questions manually (or export a SciSpace collection to CSV/RIS for the papers you have already shortlisted) rather than expecting a one-click transfer.
- Warning: Ranking signals are not identical (Elicit factors citation count and recency into its AI ranking); the top-N ordering will differ from SciSpace, so judge by relevance of the returned set, not by position-for-position parity.