Data is essential for any business today. But having access to data and using it effectively are two very different things. For instance, are you sure the picture it’s painting for you is really accurate?
If you’re working off incorrect assumptions or incomplete data that doesn’t tell the whole story, you could end up talking to the wrong people, or giving them messages that are at best irrelevant, and at worst, will actively push them away.
Your CRM and billing data may tell you, for example, that your customer is a male in their mid-50s. However, behavioral insight shows they’re into fashion, Instagram and Tiktok. Seems unlikely, until you consider that the person paying the bill isn’t the same as the one using the phone.
In this case, the chances are you’re looking at a father paying for his daughter’s phone. But if you were only using your CRM platform as a basis for your marketing messages, you’d be sending completely wrong materials to the wrong person.
This is one simple example of how incomplete data can lead you down the wrong path. So how can you avoid such errors?
The first step to achieving success starts before you ever load the raw data into your analytics platform. It’s about making sure you know what you’re looking for, and this means asking the right questions.
Do you need to understand more about behavior? Do you want to see what issues they’re having? Is there a disconnect between the CRM and behavioral data? Knowing this ensures you’re looking at the right details and focusing your analysis.
The volume of data telcos collect on their customers is vast. When you factor in billing information, CRM details, phone usage records and behavioral data such as weblogs, you’d think you have everything you need. But this isn’t necessarily the case for every customer.
There’s an old lesson taught in statistics about how you shouldn’t overlook gaps in your data. In World War II, engineers would study planes that returned from missions to determine where they should be adding additional armor. But instead of placing more protection on the parts of the planes that were riddled with bullet holes, the answer was to armor the areas where there wasn’t any damage.
Why? Because the planes with these bullet holes were still coming back. Whereas the fact there were no planes returning with damage in places like the engines doesn’t suggest they weren’t being hit there, but that any plane that was struck in this section wasn’t making it back to be studied – and therefore, these are the areas that need to be focused on.
The same principles apply to your data analytics. For instance, you may have data that shows certain users keep getting in touch to make a similar complaint. But if this only accounts for a small percentage of your user base, you might conclude their issue isn’t a widespread problem. Alternatively, if they stop calling, you may determine the problem has been fixed.
However, what about the people who don’t complain? Many people may simply stay silent until the end of their contract, then take their business elsewhere. Others may give up in frustration and, if you don’t have any complaint data to realize they’re unhappy, there may be nothing you can do to retain them.
But by looking at the data you do have and identifying where gaps lie, you can make connections, spot patterns and draw inferences that let you be proactive and address issues you may otherwise be unaware of.
When it comes to behavioral data, context matters. This means you need to look beyond the initial level of information to gain true insight into not just what your customers are doing, but why they’re doing it.
For instance, say your weblogs show someone has been browsing job boards. This could indicate they’re keen to take the next step in their career and find a higher-paying position, which may make them a potential target for an upsell. But it could also suggest they’ve just been made redundant and money is tight, in which case such an offer is unlikely to be well-received.
This is where having context and going beyond initial data points matters. By looking deeper into their data and seeing what details are there – or not – you can make sure the campaigns you’re presenting are tailored and relevant.
For example, does your job-searching customer have a history of browsing these sites every couple of years? If so, this may tell you they like to look for new opportunities, giving you an insight into their mindset and behavior. But if they’ve never looked at these sites before, or concurrently have started looking at discount code websites at the same time, this could indicate it’s not likely to be their decision to find a new job.
Knowing what data to look for and how to delve deeper to build up a complete, accurate picture of your customers is the key to success, and this still requires a human touch.
The data alone isn’t the answer. It’s useless without context and business knowledge, and being able to easily explore and question the data in different ways, draw inferences from other customers and make connections between different data points is vital if you’re to make the best use of your analytics.
Find out more about how Intent HQ can help you turn your data into valuable insight that transforms your marketing performance.
Image credit: iStockphoto/metamorworks
One of the most significant challenges facing the world today is global financial inclusion. Excitingly, recent developments in AI and ML technologies offer the potential to break this cycle and broaden access to a wealth of financial services.
Privacy is the main concern to using weblog data Telco customers reward true personalization with high ROI. At the same time, they demand a brand experience that protects their privacy. A dataset that could solve both of those issues, when used correctly, is weblogs. An incredibly underutilized asset that all telcos possess. However, concerns about […]
Intent HQ unlocks hidden customer insights that activate more effective campaign targeting and provide a better customer experience.
Data lakes have been around for eight years, and while the original concept was promising, many of the lakes have unfortunately turned into data swamps. This has prompted criticism of the concept itself.