SAASINSPECTOR

How we rate AI Research tools

Every tool starts at zero and earns points for what it does and what it costs. Here is exactly what earns them.

Last updated September 2026

The short version

  • We check each tool on 6 things a buyer cares about. Together they are worth 100 points.
  • The final score is those points out of 10. A tool with 84 points scores 8.4.
  • Everything comes from the tool's own website. We don't take payment for points.
20
25
10
20
15
10
Finding papersReading and analysing papersSystematic reviewsPricingCiting and writingAccess, teams and privacy

What earns points

1. Finding papers

20 of 100 points

Can it find the papers you need? Search across published research, answers to a research question that cite their sources, maps of connected papers and alerts when something new comes out.

Searches a database of papersup to 3
Yes+3
No0
Answers a research question from the literatureup to 2.5
Yes+2.5
No0
Map of connected papersup to 1.5
Yes+1.5
No0
Alerts for new papersup to 1.5
Yes+1.5
No0
Shows if later papers support or dispute a paperup to 1.5
Yes+1.5
No0

Each section adds up to 10, then counts for 20 of the 100.

2. Reading and analysing papers

25 of 100 points

Once you have the papers, can it help you understand them? Asking questions of a paper, summaries, pulling data out of many papers at once, and answers you can check against the text.

Ask questions of a paperup to 2
Yes+2
No0
Pulls data from many papers into a tableup to 2
Yes+2
No0
Summarises papersup to 1.5
Yes+1.5
No0
Answers show the passage they came fromup to 1.5
Yes+1.5
No0
Reads tables and figuresup to 1.5
Yes+1.5
No0
Writes a literature review for youup to 1.5
Yes+1.5
No0

Each section adds up to 10, then counts for 25 of the 100.

3. Systematic reviews

10 of 100 points

Can it run a systematic review: screening with include and exclude decisions, AI help with screening, removing duplicates, a PRISMA diagram and risk of bias?

Screening with include and exclude decisionsup to 3
Yes+3
No0
AI help with screeningup to 2
Yes+2
No0
Removes duplicate referencesup to 2
Yes+2
No0
PRISMA flow diagramup to 1.5
Yes+1.5
No0
Risk of bias assessmentup to 1.5
Yes+1.5
No0

Each section adds up to 10, then counts for 10 of the 100.

4. Pricing

20 of 100 points

What the cheapest paid plan costs, at the academic rate where the vendor publishes one. A free tool earns the most points.

Tools charge in different ways, so there is one price table for each. Only the table that matches how a tool charges counts, so a fee per use or a yearly price is never compared with a monthly plan.

Monthly price, cheapest paid plan (academic rate if offered)up to 10
Free+10
Up to $10+9
Up to $15+8
Up to $20+7
Up to $30+5.5
Up to $50+4
Up to $80+2.5
Over $80+1
Yearly price, sold by the year onlyup to 10
Free+10
Up to $120+9
Up to $180+8
Up to $240+7
Up to $360+5.5
Up to $600+4
Up to $960+2.5
Over $960+1

Each section adds up to 10, then counts for 20 of the 100.

5. Citing and writing

15 of 100 points

Does it fit how you write? Your reference manager, the export formats you need, help with writing, and add-ins for your browser and word processor.

Works with Zotero, Mendeley, EndNote and othersup to 3.5
None0
1 reference manager+2
2 reference managers+3
3 reference managers or more+3.5
Export formatsup to 2.5
None0
1 format+1
2 formats+2
3 formats or more+2.5
Helps you write and adds citationsup to 1.5
Yes+1.5
No0
Browser extensionup to 1.25
Yes+1.3
No0
Word or Google Docs add-inup to 1.25
Yes+1.3
No0

Each section adds up to 10, then counts for 15 of the 100.

6. Access, teams and privacy

10 of 100 points

Can your university or library license it, can you share projects with your team, and does it keep your uploads out of its AI training?

University or library licenceup to 3
Yes+3
No0
Shared projects with your teamup to 2
Yes+2
No0
APIup to 2
Yes+2
No0
Mobile appup to 1
Yes+1
No0
Works in languages other than Englishup to 1
Yes+1
No0
Does not train AI on your dataup to 1
Yes+1
Only if you opt out+0.5
No0

Each section adds up to 10, then counts for 10 of the 100.

When a tool doesn't say

If a tool's website doesn't answer a yes-or-no question, it gets half the points for that item. Not zero, because a missing answer isn't the same as a no, and not full marks, because silence shouldn't beat a vendor who answers honestly. For a number, like a price, it gets the typical value of the tools that do publish one. Every item scored this way is marked on the tool's scorecard.

If a tool leaves out more than a third of what we check, we don't show a score for it at all. A number built mostly on guesses wouldn't be fair to you or to the tool. Right now 23 of the 23 tools we checked have enough information to be scored.

A real example: Elicit

Here is how Elicit got its 7.8.

SectionPoints
Finding papers14 of 20
Reading and analysing papers25 of 25
Systematic reviews8.5 of 10
Pricing14 of 20
Citing and writing8.6 of 15
Access, teams and privacy7.5 of 10
Total77.6 of 100 = 7.8 out of 10

See every fact behind Elicit's score.

What the score doesn't tell you

We score what a tool offers and what it costs, as its own website describes it. That's not a hands-on test. The score can't tell you whether a tool is easy to use, whether its data is accurate, or whether support answers quickly. The review covers that.

See the AI Research tools ranked or read how we review tools in general.