SIC codes are a powerful data analysis tool that enables you to immediately understand the purpose of any of the UK’s six million businesses.
In this article, we’ll provide an overview of what SIC codes are and explain how Red Flag Alert’s machine learning technology unlocks their value, before providing some examples of how this could help your business.
All UK companies are required to select a SIC code when they first register with Companies House. They are used to classify the type of economic activity which the company undertakes. This data can also be useful for anyone wanting to understand more about the business.
An example would be 95 – Repair of computers and personal and household goods. This provides a useful starting point to understanding the business, but, like most SIC codes, it is vague and could even be misleading.
Some registries or organisations try to enrich this information with more in-depth detail, but these can lead to new difficulties. For example, telephone interviews can provide greater granularity; however, they are also prone to inconsistencies and human error.
Keyword assignment models can also be used; however, these don’t tend to recognise words used out of context and are, therefore, hit and miss.
Here at Red Flag Alert, we have invested in the latest machine learning technology to revolutionise the way we assign SIC codes. This has resulted in highly accurate codes that can be relied on when making decisions for your business.
So how does it work?
We use several ‘big data’ techniques to process company information and assign them a more precise, more accurate SIC code.
Let’s take our example 95 – Repair of computers and personal and household goods again.
Looking up the same company on our SIC database returns a result of 95.23 – Repair of footwear and leather goods.
This gives us much more information, and we now know that the company repairs shoes and leather rather than computers – two very different businesses.
We generate this more detailed categorisation using three main processes:
When adding a new business, we input the usual propriety business information into our database, before feeding it into a word cloud. The artificial intelligence system then conducts the semantic search, comparing the word cloud information against a database of contextual meanings.
Red Flag Alert’s own SIC experts will also be inputting into this, carrying out extensive QA on a random sample of SICs, which effectively ‘teaches’ the machine by giving it human-generated reference points.
With the information of nearly six million companies already in our database, the word cloud already contains billions of semantic connections, and this is continually growing, meaning that the artificial intelligence can assign accurate, granular SIC codes both now and in the future.
Customers will also gain a more precise, granular picture of what a business does through multiple SIC codes, each of which provides a different angle on the company’s operations.
They can also be confident in basing business decisions on our SIC codes. Each is assigned an assignment score, so we know if it is within an acceptable threshold.
Advanced SIC codes provide an edge in a range of different business lines. They include:
Using SIC codes to acquire more intelligent and granular information means you are giving yourself a head start on your competition.
Whether you’re looking to grow your sales or boost your risk analysis, SIC codes help you to gain a clearer picture of the companies that you are doing business with.
Discover how Red Flag Alert’s experienced team can help you mitigate risk and protect your business. Why not get a free trial today and see how Red Flag Alert can help your business?