Working out loud

Identifying outliers

We’ve added the ability to remove likely anomalies from your analysis.

When performing any analysis, it is important to consider outliers. That could be considering whether a company’s main activity falls within the sector that you are analysing, for example.

It is also important to consider financial outliers or anomalies.

We have recently introduced a new feature that will make it easier to identify where companies’ accounts are misrepresented. This is part of our continuous commitment to providing the highest quality data.

Outliers affect less than 0.05% of our companies but their impact, by nature, can be large. Identifying possible outliers will allow our users to review and remove the very small number of companies that have bad data.

Finding the number of outliers

The first place you’ll notice this feature is in ANALYSE. In ANALYSE you’ll receive information on the number of likely anomalies in a list.

Inspecting outliers

We’ve also added new company filters in ANALYSE and EXPLORE.

“Only outliers” allows you to inspect the list of outliers. The best way to inspect the list is by viewing it in EXPLORE.

TIP: From ANALYSE you can select “View List” with the view outlier filter applied to see the list in EXPLORE (and vice-versa!).

Within EXPLORE, you can view the reasons for each company being identified as having at least one year of misrepresented accounts.

For example, in the FINANCIALS tab of company 00898353, POSTLIP PENSIONS LIMITED reports having 265 employees with £1000 in assets. This is very unlikely to be true. Therefore, their financial anomaly years are highlighted.

Excluding outliers

Now choose “Exclude outliers” from the companies filter and the affected companies will be removed from your explore list, or your analysis in ANALYSE.

If you want to read more about why we developed this feature and the types of anomalies that occur, or have other questions about our data, please take a look at our Knowledge Base.

While things like this seem obvious from the outside, it’s been a massive undertaking. But as I already mentioned, spotting these outliers is pivotal in getting accurate statistics.

Always more data updates to come, so keep an eye on the blog! And while you’re here, why not sign up for a free trial?

About the author