Data prices · Chemistry & materials
What AI labs pay for chemistry data
Labs building science models pay for complete lab notebooks: conditions, yields, analytical results and the failed runs that papers leave out. There are no public prices for chemistry notebook data, but published research shows why failed experiments are worth paying for.
What buyers want
Records your team already keeps
01
Full notebooks
Reagents, conditions, observations and results, including runs that didn't work.
02
Linked analysis
HPLC, NMR and QC results tied to the experiment that produced them.
03
Scale-up history
What changed between bench and plant, and why.
Price signals
What public deals and research show
89%
Success rate of a model trained on failed "dark reactions" from lab notebooks when predicting new product formation. It outperformed traditional human strategies.
Raccuglia et al., Nature 2016$200-$2,000
Typical price per checkable task. A notebook entry with a known outcome can become one.
Epoch AI, 2026We found no published price for chemistry notebook or ELN data. Large pharma companies have often preferred to train shared models without moving their data at all, as in the MELLODDY project, which tells you how carefully this data is guarded.
The wider market
$203M
Data licensing contracts Reddit disclosed in early 2024, with terms of two to three years.
TechCrunch$250M+
Reported value of News Corp's five-year content deal with OpenAI, including cash and credits.
Press Gazette$10k-$100k
Per-company payouts for shut-down startups' Slack messages, tickets and email archives, across nearly 100 deals.
Forbes, April 2026$200-$2,000
Typical price per reinforcement learning task, according to Epoch AI's interviews with people who build them.
Epoch AI, 2026Pricing factors
What moves the price
- Failed runs included
- A model trained on archived failed reactions predicted successful conditions with an 89% success rate, beating traditional human strategies.
- Structure
- Conditions recorded as fields, not only free text, are far easier to use.
- Rarity
- Chemistries and materials few others work on are worth more.
- Years of history
- Longer runs of records show change over time and how outcomes played out, which a single year can't.
- Rights you can prove
- Buyers pay for data you own or are allowed to license. Data held for customers without their permission won't sell, so our calculator cuts the estimate when rights are unclear.
Our calculator
How we estimate chemistry data
Our indicative model starts from a base value per system for about five years of history at an 11-50 person company, then applies a sector weight of 1.6x, your history, team size and rights. It is an estimate we confirm on a call, not a market price.
| System | Base value |
|---|---|
| Lab notebooks & LIMS | $60k |
| Docs, wikis & SOPs | $10k |
| ERP & operations | $25k |
Recurring
Paid again every month
Active labs log new experiments every week. On a refresh term, each month's notebook entries go out as a new batch and pay again.
How recurring revenue worksRules that apply
Customer projects and client-owned results need the client's permission. Pharma companies trained a shared model by federated learning in MELLODDY precisely so their own data never left their control.
Source: IHI, MELLODDYValuation
What is your data worth?
No access to your systems needed. We confirm the number on a 30-minute call.
Questions
Chemistry data, answered
Why would anyone pay for failed experiments?
What about client projects?
Do you need our ELN login?
Keep reading
Selling chemistry dataSelling lab notebook data to AI companies
Lab notebooks sell when they record full conditions and failed runs. How labs value them, what has to be removed first, and how a license works.
6 min read
What makes data valuable to AI labs
Labs pay most for records they can't scrape or generate: real decisions, real outcomes, long history and fresh data, with clear rights to use them.
5 min read
Episodes, records and environments: the three things AI labs buy
Records are de-identified rows, episodes are human actions with outcomes in order, and environments are working replicas. Each sells at a different price.
6 min read
Prices in other industries