01Available
ISA decisions
- Observations
- 8,750
- Sessions
- 3,700
- Visible providers
- 22
- Latest period
- 2026-08
Coverage & methodology
How the data is collected,
How the data is collected,
bounded and published.
This guide describes the observation sources, analytical limits and publication rules that govern all data on this site. Read it before drawing conclusions from any single result.
What is currently covered
Three observation datasets, covering distinct planning decisions.
The cards below show the sources currently active in the repository. Observation counts and provider tallies are computed server-side; the browser receives only approved aggregate output.
02Available
Retirement & drawdown
- Observations
- 4,450
- Sessions
- 1,750
- Visible providers
- 14
- Latest period
- 2026-08
03Available
SIPP decisions
- Observations
- 2,000
- Sessions
- 1,000
- Visible providers
- 8
- Latest period
- 2026-08
Analytical terms
Key terms used throughout the repository.
These definitions apply consistently across all public datasets and provider profiles. Read them before comparing figures across dimensions or time periods.
Observation
One completed calculator decision in a source table. It is not necessarily one unique person or household.
Session
A technical source reference used to produce combined counts. Session identifiers are never included in public data.
Provider profile
A public summary for a provider appearing often enough in a dataset. It is not a rating, recommendation, ranking or estimate of market share.
Typical value
Usually the median in a published group. A value appears only when there are enough valid numbers to calculate it safely.
Publication rules
The standards that govern public disclosure.
These are methodological constraints, not fine print. They ensure the repository is analytically useful without exposing granular or potentially re-identifiable patterns.
Minimum group size
A group needs at least 50 observations before it is included in a public chart or table. Smaller groups are omitted, not estimated.
No personal data in the dataset
No personally identifiable information is ever collected or processed in the decision data covered by this repository. Names, emails, addresses, IPs, device IDs, source identifiers and session IDs are not part of the analytical dataset, and public pages return combined results only.
Not personal advice
Patterns in the data can inform a question, but they cannot tell an individual what financial choice is right for them.
Research enquiries
Get in touch →
Working with the data? Talk to the team.
We can clarify what the public data covers and discuss research questions without opening a raw-data endpoint.