Hello everyone
We're currently working on a customer segmentation project with an external consultant and we anticipate using Attributes to hold the segment information, and to use Manage Attributes to apply the relevant segment to each customer.
Does anyone have any experience of doing this and if so, do you have any recommendations of how to approach it? We're anticipation having 6-7 segments, though whether these are to be then sub-divided, we await the presentation!
Also, this is a process that we expect to do again in a year (and subsequent years) and wondered if anyone had done the same thing and whether you added another attribute for each year, so you can track any movement between segments?
Any thoughts on this subject are very welcome
Thanks, Christopher
Hi Chris,
We already hold our segmentation in Tessitura. We have 7 segments plus sub categorisation.
We use Constituency for the main segmentation as this allowed use to se a start and end date for the segmentation which made it easier for updating as we set each constituency for 12 months then update it just before it ends, used the start date as first performance date that way we only need to update less than 10% of our cusromers every month. Also means we have a history of which segment they were assigned to at a particular time, if their buying habits change.
Our sub-categorisation is held in an attribute (actually multple attributes due to the fact that this is essentially a life stage categorisation and so we need to know what they were previously categorised as before assigning the new one).
We also stored our variables as attributes, these get updated every few months and are then used when the customer's segmentation needs updating.
Hope that helps, if you need any more info, please get in touch.
Mark
Hi Christopher,
We hold our segments as attributes and have a naming convention so a new attribute is created for each update of the segments. For example, SegmentationStratford(Oct09) is our latest segment attribute and the values sit against that (e.g. Regular, Ebookers etc.). We can then track back using lists and extractions to see the movement between segments as each iteration of the segments has it's own attribute.
We can also select people based on having remained in a given segment (e.g. Regulars for the last four segmentations). We conduct our segmentations quarterly and have found this system works really well.
Let me know if you would like any more detail.
Mary