One of the most critical components of web analytics and conversion rate optimization is interpreting data, analyzing it, and turning it around like a Rubik’s cube. For instance, you can rotate and arrange the traditional 3x3x3 cube in 43 quintillion ways – 43,252,003,274,489,856,000 to be precise. Similarly, you can build billions of models from your data. However, only one of these models will give you the best conversion rate, and this is the elusive Holy Grail you must find to maximize your efforts.
Let me illustrate this with another example. As you can see in the image below, there are 3 expressways you can take to go from Chicago to Wyoming. The shortest route is the 196, but if you take a detour and go via 131, you might end up adding an hour to your journey. There are a couple of other routes to get there too, but both will take longer.
Image Source: Google Maps screenshot
In the same way, any of your conversion tactics can prove lucrative and get you to your goal, but one will get you there faster than the others.
So how do you find the perfect, elusive model to reach the optimum conversion rate? The answer lies in data democratization, data decentralization, and data transparency. Let’s dig a bit further into the whys and hows of revolutionary data processes so you can play around with the data you’ve collected and find your conversion sweet-spot.
We have more data than ever before, sometimes more than we know what to do with. If you were to narrow down data and scrutinize bounce rates on Google Analytics, you would probably start with entry and exit pages, then move on to user flow, keywords, location, language, device, browser compatibility, site speed, and more.
This is just one issue (bounces) and one data source (Google Analytics). Now imagine you add legacy data, POS transactions, social media conversations, survey results or feedback collected from trade events, and make inferences from the pool of data available: even if you are an insatiable data glutton, you’ll be quickly overwhelmed!
It is inappropriate, impractical, and immature to restrict data to the elite few, whether they are CEOs or domain-specialized data scientists, and expect them to make sense out of it for everyone. This is why 2016 has been heralded as the year of data democratization. Data democratization ensures everyone in an organization has access to data, and is therefore in a better position to make decisions.
Finish Line, an athletic apparel and footwear retailer with nearly 700 physical and online outlets, uses POS data, loyalty data, social streams, and beacon data to improve one-on-one communication and up-sell to customers. The results were impressive: Stephanie Bleymaier, Director of Digital Personalization and Loyalty, reported a 50% increase in email open rates and a 30% rise in return on social media ad spend. Guess where that would have taken conversions?
Related Class: Leverage Social & Customer Data for Email Relevancy
Other departments were able to “increase their efficiency...by tapping into the data pool,” Stephanie said. The first priority for any decent collaboration and data management tool is making data available to all users, ensuring healthy debate, consensus, leading to quicker, more informed decision-making.
The idea of data democratization may cause worries over security. But to protect sensitive data, you can use a sophisticated collaboration tool that offers secure file sharing, access control and group or individual level permissions, ensuring data doesn’t fall into the wrong hands. For instance, campaign tracking and project management tool WorkZone allows users to share sensitive data and plans securely with a particular team, giving each person access to just the right amount of information.
Image Source: WorkZone screenshot
With this tool, users can send emails, comment, make changes or record meeting notes to data resources that are relevant to them. WorkZone also automates the approval workflow, sending requests to managers, tracking responses, and recording access. This way the entire team stays informed, and you have an organized, time-stamped record of all activity.
As we saw above, data democratization bought a significant jump in email open rates for Finish Line, but democratizing doesn’t happen on its own. You need to make your data more intelligent and intuitive for any layman to understand. A standard practice of data democratization is to collect data from all touch points and make it available to each and every user. This is known as data centralization.
From the conversion point of view, data centralization is an exercise in profiling every customer, by collecting data on their preferences and behavior from CRM, point of sale, logistics, customer services incident management, and other systems, integrating all of it in one common data bank. When you enter a particular query about the customer, their data is processed, de-duplicated, and cleaned to give you accurate search results in a simple, transparent and aesthetically pleasing format.
This is how data is crunched, in a cost-effective way in large organizations.
This is the way things have been done up until now.
This is a bad way to democratize data.
Data centralization makes your analytics sluggish and its results are unintuitive. Here’s an analogy: say you are a data scientist or a user, your furniture is your data, and your house is a storage unit. Data centralization means you put all your furniture in one single unit or a room so you have a single point of access. Now imagine finding and pulling something out of such a room.
Data decentralization means letting your data stay where it is, so each unit is able to localize, process, and analyze data in an agile manner. This way, you go directly to the kitchen if you need to access the chopping table, and consequently, get the job done quickly.
Telecom major Vodafone successfully kept their decentralized systems in operation and created a unified view of customer data for their customer service agents, while minimizing data replication by using Denodo’s data virtualization platform. Denodo added a data virtualization layer to legacy systems, liberating data by keeping it where it was, but allowing users to get a complete, updated view of customer data, even though it was scattered across disparate sources.
Image Source: Denodo
Vodafone was able to reduce average service response time from 6 to 2 minutes. They also found better upselling and cross-selling opportunities, retained more customers, and stretched the usability of existing data infrastructure as a result.
Related Class: Drowning in Data: How to Effectively Leverage Web Analytics
The last piece of the puzzle is to increase the clarity and comprehensibility of data. When I say “data transparency,” I mean inside and outside the organization. In a bold move, Omniconvert (formerly Marketizator) announced a product revamp strategy that allows marketers and customers to decide on the features and capabilities for what they call “the first democratic CRO software.”
Source: Omniconvert screenshot
Perhaps Omniconvert’s strategy is risky. But it is also rewarding. It allows organizations to build products while maintaining close contact with their communities, ensuring customers get what they want. With this master stroke, Omniconvert has banded experts and users together to create a successful product that has the features and agility of an open source platform while maintaining the reliability and robustness of proprietary software.
Valentin Radu, CEO and founder at Omniconvert is optimistic about the move since “everyone can publicly see the priorities of the features as they are voted for.” The whole voting system is completely transparent, as every planned update with new features is publicly shown on the site. There’s also video streaming of all the kick-off meetings, making the whole process more engaging and open-sourced.
Over to You
These are just a few ways you can liberate data and make it more accessible to users, getting you to your goals faster. How does your organization use data in creative and intelligent ways to increase conversions? Let’s hear your stories in the comments below!
Tracy Vides is a content strategist and researcher who gives small business and entrepreneurs marketing and social media advice. Tracy is also a prolific blogger - her posts are featured on Engadget, She Owns It and Usability Geek. Connect with her on Twitter @TracyVides for a chat anytime!
Want to learn more about any of the subjects mentioned above? Here are some relevant classes: Leverage Social & Customer Data for Email Relevancy, Drowning in Data: How to Effectively Leverage Web Analytics, How to Create a Data-Driven Culture