Clustering is a data analysis technique used in organisations that involves grouping similar datapoints based on certain features. Clustering algorithms aim to organise datapoints into different clusters (or groups) where the data are like each other as opposed to the other clusters.

Organisations can use clustering, or more generally known as segmenting, to help shape their marketing strategies and better understand their customer base. Typically it refers to the practice of dividing the customer base into distinct groups based on a specific characteristic, such as demographic, purchasing potential, size, or region. For example, something as simple as labelling a customer as "B2B" or "B2C" would be a form of segmentation. You could take this further by then classifying them into groups based on their average order values.

K-Means is a clustering method in Machine Learning. Machine Learning is a subset of Artificial Intelligence, and focuses on the development of algorithms that learn from the data and improve over time. We've used this clustering method to group our database of customers into several distinct “AI Clusters”. By using an AI algorithm, you can reveal hidden relationships in the underlying data that would otherwise not be so apparent.

We use these AI Clusters as the basis for the benchmarking. Once you're assigned a cluster of similar businesses to your own, we determine the different quartiles for each period of each of the metrics in the Growth Engine, allowing you to see how well you're doing compared to similar businesses. By understanding how well you're performing compared to other, similar organisations, you're then able to determine whether or not your performance is above, below, or just about where it should be. For example, you may be experiencing revenue growth of 3% consistently, which on its own is a positive result. However, you might find that your growth is in the lower quartile of your cluster, so it may become an area of focus.

There is also value in understanding how you compare to companies within your region or sector. We know that many businesses provide services or products that fit a particular niche or a more specialist level of quality, so you may feel that a direct comparison is not as applicable. But there may still be insightful information you can glean. For example, a wholesaler of a unique coffee blend could find great value in understanding what the wider sector is experiencing. Not only that but then being able to analyse their specific region's trends could help an organisation determine whether they're adequately achieving their targets.

The algorithm looks at a few different features, including your previous Growth Engine figures; so, if you only subscribed to the CRM recently, we're unable to accurately predict you into a cluster as we won't have enough data just yet. However, as you use the CRM, we'll use a different AI algorithm to try and estimate what some of the missing data might look like so that we can fit you to the most appropriate cluster. Then, you can start getting some value from this truly unique feature of the CRM! Alongside the AI Clusters, you can filter the benchmark capability by sector or region, allowing for further analysis.