What could your research become?
Explore where your work could matter beyond your immediate field — from emerging technical capabilities to commercial applications and potential businesses.
Where the investigation went outside your papers, and why.
Your papers are the seeds. When a question needs a cost benchmark, a regulator's test method or a market's loss figures, the investigation goes and gets it, records why it went, and counts what survived the checks. One such source from each published investigation.
- From the batteries investigationIndustry whitepaper
Cost Projections for Utility-Scale Battery Storage: 2025 Update (opens in a new tab)
Why the investigation went there. NREL 2025 utility-scale battery cost projections: the newest public $/kWh benchmark for grid storage.
4 claims from it survived the checks and are cited in the report.
See how it was found - From the oil & gas investigationWeb source
Alternative Test Method (MATM-010) (opens in a new tab)
Why the investigation went there. EPA-approved alternative test method for a commercial continuous monitor — the performance specs a continuous system must meet.
5 claims from it survived the checks and are cited in the report.
See how it was found - From the finance investigationIndustry whitepaper
PS25/5 APP scams reimbursement requirement consolidated policy statement (opens in a new tab)
Why the investigation went there. UK regulator's APP scam reimbursement rules (liability split between sending and receiving PSPs, cap): the regulatory driver that makes scam and mule detection a paid-for problem.
6 claims from it survived the checks and are cited in the report.
See how it was found
Find where your expertise could matter outside the field where you developed it.
Researchers often understand their immediate literature extraordinarily well. They may know far less about distant industries with structurally similar problems.
Between Papers searches beyond familiar disciplinary boundaries, and shows which sources it went to find and why.
The test of a good investigation here is that you find yourself saying: I knew what my research did. I hadn't realized where else it might matter.
Named buyers, not 'industry'
Each opportunity names who would pay and how they buy. The question a researcher rarely gets answered.
Receiving-side mule account detection for UK payment firms
- Who buys
- Fraud and financial-crime leads at UK banks, EMIs and payment institutions that now pay half of every APP scam reimbursement as receiving firms.
- What to build
- A mule-risk score for incoming Faster Payments accounts that combines account behaviour with network features and explains each flag, so the receiving firm can hold or freeze funds before they move on.
- Why now
- Since October 2024 receiving firms fund half of reimbursements up to £85,000, a direct cost for hosting mule accounts. A bank's mule model with explanations already outperformed its rules in a live pilot.
Main risk. Fraud may pivot to international payments outside the scheme, shrinking the liability that motivates buyers.
Aging-aware operating software for storage fleets, ready for sodium-ion
- Who buys
- Grid battery operators and home storage fleet owners, bought as a per-MWh software subscription or bundled by system integrators.
- What to build
- Software that estimates each battery's health from normal operating data, prices aging into trading decisions, and answers operator questions automatically. Support lithium iron phosphate and sodium-ion cells from day one.
- Why now
- In one simulation, misjudging aging cost cut trading profit by 30-50%. In a 30-query test, an AI assistant cut per-query cost by about 98% versus manual expert analysis. Sodium-ion health was estimated from partial charges in a lab dataset.
Main risk. Cell makers and integrators lock data inside their own management systems and offer similar analytics free.
Alert-to-repair service that turns satellite detections into fixes
- Who buys
- Assumed buyers: national oil companies and small independents in emerging economies, with contracts funded by governments, development banks or climate funds.
- What to build
- A response desk for national oil companies and regulators: triage satellite alerts, send local teams to find the source within days, organise the repair, and confirm the fix with a follow-up satellite pass. A possible pricing model is per verified tonne avoided.
- Why now
- Alert systems now notify thousands of events but most go unanswered; averaged analyses find abating high emitters a net saving in most high-emitting countries, and the IEA sees public funds for capacity building.
Main risk. Operators and host governments do not grant access or act, as the low reply rate suggests.
The cheapest experiment
Every combined idea ends with how to test it. This is the line a spinout conversation starts from.
- From the batteries investigation
Local waste-wood hard carbon to fix Europe's sodium-ion cost gap
How to test it. Make kilogram batches from local waste wood, build pouch cells against a commercial cathode, and cost the process against imported hard carbon.
Where it breaks. Very high carbonization temperatures are energy-hungry, results are from half-cells or small full cells, and no cost figures exist.
See it in the investigation - From the oil & gas investigation
Randomised trials of satellite alert protocols to price an alert
How to test it. Randomise a few hundred repeat emitters into notification arms and compare re-detection rates over six months of satellite passes.
Where it breaks. Satellites see only large emitters and revisit rarely, so outcomes are noisy and small sites cannot be tracked.
See it in the investigation - From the finance investigation
Scambaiting bots as a fresh label source for mule-account models
How to test it. Match a month of harvested account numbers against a bank's accounts; measure how many were later confirmed as mules and how many days earlier they were flagged than police-derived labels.
Where it breaks. Harvested accounts may be few, short-lived or already burned; only 48.7% of engagements got any reply, and label provenance may not satisfy bank legal teams.
See it in the investigation
How it works
- 01
Supply the research
Upload 25–100 papers and tell Between Papers what you're trying to understand.
- 02
Build the evidence
Technical findings, experimental conditions, limitations, agreements, and contradictions are extracted and connected across the corpus. Each claim is checked against the passage it came from.
- 03
Investigate what they enable
Findings are connected into emerging technical capabilities, including potentially useful connections from adjacent fields and, where it helps, sources from outside your corpus.
- 04
Explore commercial implications
Capabilities become potential applications and business opportunities, while evidence, assumptions, and uncertainty remain visible.
Start free: the agent reads your seed papers and gives you the answer and the first findings. Pay only to continue the same investigation to the complete report.
Questions
Why not just upload the papers to ChatGPT or Claude?
General-purpose models can read and summarize research. Between Papers imposes a structured investigation process designed specifically around cross-paper evidence, technical conditions, provenance, emerging capabilities, and commercial implications.
Does it search outside my corpus?
When useful, the investigation can use targeted external research to test assumptions, investigate adjacent fields, and assess commercial context.
Can I verify the conclusions?
Yes. Important conclusions can be traced back through supporting findings to source passages.
Not your job? Find your next product in the research. · See what the research actually makes possible.
Find where your expertise could matter outside the field where you developed it.
Start free with your own papers: the answer and the first findings cost nothing.
The flagship: cheaper batteries for homes and the grid, 25 seed papers, 396 sources looked at.