Effective marketing is based on research, analysing data and tracking performance metrics. But if you’re not aware of the ‘giraffe effect’, you could be driving your business toward disaster.
All the answers to your unique business lifestage questions
It seems strange that something as benign sounding as the ‘giraffe effect’ can have such cataclysmic consequences for a business’s growth aspirations.
Simply put, a giraffe in your data misleads you into wrong decisions and results in missed opportunities.
Giraffes can be described as portions of data that stick out from and dominate the rest of the data.
The effect they have is hiding important insights that would result in a very different conclusion and course of action. Because they stick out the way they do, they also skew the size of other insights which, by themselves, would normally be given more attention.
How this applies to growing your business
Let’s use a customer analytics giraffe to illustrate the disastrous consequences of being misled.
If you’re looking to improve customer acquisition efforts by focusing on the most lucrative customer segments, you might look to customer gender as a first stop.
Top level aggregation of data reveals your male customers have a higher lifetime value (LTV) than your female customers.
The assumption from this data would be to focus more resources on upping male customers than female, right? Well, no. Viewed in the isolated context of your business, you’d think that’s the case, but zoom out to view your industry and look what happens.
As it turns out, with greater context it appears your initial data-set is an anomaly compared to the industry at large where females, in fact, have much higher lifetime value than males.
The next time you’re analysing data for marketing purposes, make sure you take the following into consideration to identify and work around potential giraffes:
- Ensure your SEO data eliminates all traffic due to searches including your brand’s name.
- Have you made provision for one time purchases potentially concealing important insights on repeat customers?
- Have you dug past the obvious data conclusions to identify and work around dominant and potentially misleading data of high-level aggregated views?
- Try to avoid viewing data in isolation – context is key.
- If you have context to your data, don’t go looking for unicorns until you’ve run out of ponies. Metaphor to animal oriented? Provided you’ve got sufficient context, don’t go looking for answers you want to see when you’ve got them right in front of you. Collect your data, analyse it, and act accordingly.