Tech Nation: What it is and how we made it

05 February 2015 DueDil Team

Just how important is tech to the UK economy?

Today (5th February 2015) sees the release of Tech Nation, an attempt to map and analyse all the digital businesses in Britain. Partnering with Tech City, we’ve identified 47,220 companies who have a total turnover of £119bn and employ 7.5% of the British workforce. The in-depth analysis, powered by DueDil’s advanced search, reveals for the first time the strength of the UK’s tech economy in driving Britain’s national growth.

We believe this is the most accurate projection of the tech industry today. It’s not yet perfect, but this project is a starting point in trying to measure this sector.

What is it and how do I get my hands on it?

The best way is to visit www.duedil.com/technation/2015 to start digging into the data on our site.

You can also use advanced search to slice, dice and create reports as you choose and download the key report.

The data is live and constantly updated to paint the most accurate picture.

But, if you’re looking for some headline figures, here are some from the research for your delectation.

  • Digital job growth is predicted to outperform all other occupation categories by 2020
  • 1.46 million people – 7.5% of the entire British workforce – are currently employed in the digital industries across the UK
  • 50% of digital companies in the UK have been founded since 2008
  • 74% of digital businesses in the UK operate outside of London
  • 77% of digital businesses say they benefit from access to a network of entrepreneurs or clusters to interact with and to share ideas
  • 90% of the UK’s digital businesses expect their revenue to rise in 2015
  • The UK’s fastest growing tech clusters are: Greater Manchester, Liverpool, Belfast, Inner London, Sheffield, Bournemouth, South Wales, Brighton & Hove

How did you make it?

To make it happen, we drew on these main data sources:

1.Company websites. We drew on DueDil’s 2.7 million company-to-domain matches for 1 million+ companies.
2.AngelList data on UK digital companies.
3.CrunchBase data on UK digital companies.
4.Information was extracted from Wikipedia monthly data dumps, specifically to identify famous companies.
5.We also added the MTM survey companies.
6.We used the 2003 and 2007 Standard Industrial Classification of Economic Activities (SIC) system to include companies that self-identify as “digital”:

  • 7221 (2003) – Publishing of Software
  • 3002 – Manufacture computers & process equipment
  • 5821 (2007) – Publishing of Computer Games
  • 5829 – Other Software Publishing
  • 62011 – Ready-made Interactive Leisure And Entertainment Software Development
  • 62012 – Business And Domestic Software Development
  • 6311 – Data Processing, Hosting And Related Activities
  • 6312 – Web Portals
  • Information was extracted from Wikipedia monthly data dumps, specifically to identify famous companies.

Previous studies have been limited by relying exclusively upon SIC codes, which as we have mentioned aren’t accurate enough for robust analysis on their own.

We next extracted keywords from our core data sources. (Within unstructured text, we applied natural language processing to identify candidate keywords.)

We then conducted a classification exercise in which 2,000 companies were independently tagged by project partners as digital, non-digital or ambiguous candidates. The results of this exercise were used to create a machine-learning training set. (The training set was tested against 500 non-digital companies.)

This training set was then used to identify “digital” companies from among the wider universe of UK-registered companies, using the data sources outlined above.

As a verification stage, the crowdsourcing was used to identify and remove anomalous companies. Within this stage, and to ensure we had no poor-performing assessors, we asked multiple people to tag individual entities.

Finally, a keyword-clustering stage was implemented to enrich the keyword database.

This final database was used to show the companies within Tech Nation, and their location (based on registered and trading address, where available).

Got any questions?

We’re happy to help. Drop us a line.

If you’re a company looking to unlock our data, call us on +44 (0) 203 137 8490 or send us the message at enterprise@duedil.com.

Press enquiries should be sent to press AT duedil.com