How do we compare?

How do we compare?

We're often asked how we compare to other device detection suppliers, such as DeviceAtlas and ScientiaMobile. Using industry-defined key metrics, we've compiled this handy comparison table. These figures are constantly updating, so be sure to refer to the footnotes.

Key metrics 51Degrees DeviceAtlas ScientiaMobile /WURFL TM
Prebid modules .js, Go, Server Java23 None .js
Device combinations 2,876,6541 Unpublished Unpublished
Active web device profiles 46,5022 Unpublished Unpublished
All hardware device profiles 87,7023 >90,0004 >105,0005
TAC coverage 98.31%6 Unpublished Not available
Number of unique active TAC 72,145 Unpublished Not available
Duplicate TACs identified 2.18% Unpublished Not available
Accuracy >99.9%7 99.9%8 >99%9
Data sources Real devices, vendors, and
optional anonymous global
usage sharing7
Multi-sourced device data10 Unpublished
Multiple environments Indexed for web, native apps,
and mobile operator environments11
Indexed for web, native apps,
and mobile operator environments8
Unpublished
File format Binary JSON12 XML13
Confidence indicators - avoids false positives Yes14 No Yes15
Secure on-premise solution Yes Yes Yes
Open-source Yes16 No No
Regular expressions No Yes Yes17
Transparent pricing Yes18 No No
ISO Accreditations 9001 & 2700119 None None
Free trial with complete dataset access Yes Yes No20
Frequency of data file updates Daily for both cloud and on-premise Daily for on-premise, weekly for cloud21 Weekly22
  Start Free Trial DeviceAtlas WURFL

All competitor information accurate as of 4 December 2024 unless otherwise stated. We aim to keep the table up to date, however, if you see any inaccuracies or outdated information, please email us at marketing@51degrees.com.

  1. Combinations of hardware, operating system, and web browser that are included in the training data.

  2. The number of profiles that relate to devices that have been active on the web during the past 12 months.

  3. The number of devices contained within 51Degrees including those that are either inactive on the web or do not connect to the web (i.e. IoT devices).

  4. Source: DeviceAtlas About our data.

  5. Source: ScientiaMobile.

  6. Our database covers 98.31% of active devices (all that are currently significant and verifiable in the real world).

  7. Real devices contributed by the 51Degrees Contributor. 51Degrees data processing system contains hundreds of millions of unique HTTP headers including crawlers, apps and web browsers. Less than 1.75% of the total available headers form a training data set which is used with machine learning techniques to achieve 99.9% or greater device detection accuracy when compared to the full data set.

  8. Source: DeviceAtlas Why DeviceAtlas

  9. Source: ScientiaMobile Accuracy of our Device Intelligence.

  10. Source: DeviceAtlas: "The multiple sources used by DeviceAtlas derive from various agreements with device manufacturers and Mobile Network Operators, as well as client and user contributions".

  11. Read "51Degrees Announces Native App Device Detection Capability" for more information.

  12. Source: DeviceAtlas Device Data (JSON)

  13. Source: ScientiaMobile WURFL Schema

  14. We allow developers to choose whether they want to receive false positives or not. Refer to our False Positive feature documentation for details on how to do this.

  15. Source: ScientiaMobile User Agent Parser – WURFL

  16. See the 51Degrees Github repository.

  17. Source: ScientiaMobile's Device Detection e-Book

  18. See the 51Degrees pricing for more information.

  19. 51Degrees became the first commercial open source company in the device detection game to achieve a duo of ISO 9001 and 27001 standards for quality and information security management. Read more about how and why we embraced ISO.

  20. Source: ScientiaMobile Free Trial has a "device data snapshot file"

  21. Source: DeviceAtlas Pricing

  22. Source: ScientiaMobile "weekly XML file updates"

  23. Prebid: 51Degrees are a member of Prebid with several partner modules, whereas WURFL just have the one community module - Documentation can be found here: Prebid.js, Prebid Server Go & Prebid Server Java