Every big company continuously tries to create relevancy by differentiating itself from the competition. Big data offers tremendous chances to do so, and therefore many companies choose to differentiate through data. They collect, process and distribute data for analytics and construct a clear view of the customer. This 360 degree customer view helps companies to define the customers value to the business. This can then be the input for new campaigns, for instance to increase customer loyalty or to prevent churn.
To create this single customer view, many companies are turning to data warehousing. Throughout the entire company, there are several systems collecting massive amounts of data, such as the ERP, CRM, order process management, etc. All these tools deliver data to a data warehouse, for analysing and optimising operations and customer journeys. Many companies use data warehousing solutions by Teradata, Oracle, IBM or Microsoft for this cause. However, when it comes to delivering web data to the warehouse environment, many companies tend to struggle.
Web Analytics Tools
A common practise is the following situation: within a company a massive part of the customer journey takes place online on the website. The data from the website are however collected and processed by a web analytics solution, like Adobe Analytics, Webtrends, IBM Analytics, Webtrekk or Google Analytics. In order to map, analyse and optimise the customer journey, this data has to be extracted from the web analytics solutions and delivered to the data warehouse, just like all other customer data. But the previously mentioned analytics tools are not designed to do so, making this way of working far from optimal. Heres why.
Data Models
Web analytics tools are meant to deliver reports and clear visualisations, as that is what marketeers want to see. To deliver these reports quickly and easily, the tools store the collected web data in predefined data models. This is ideal within the walls of the web analytics data silo, but it makes the data unusable for other causes because you cannot take it out of the data model properly without losing or polluting your data. But there are several more reasons why web analytics tools are not ideal to use within a business intelligence environment where all data is delivered to a data warehouse:
1. Processing Time
Most web analytics tools need a day or more to process data. And extracting data from the tool will take even more time, making quick responses impossible and leaving customers unsatisfied.
2. Affected Data Quality
If you manage to extract data from web analytics tools, chances are that your data quality is affected due to pollution or missing data.
3. Poor Flexibility
To really optimise your customer journeys, you want to stay flexible when it comes to the data you collect. Web analytics tools do not offer this flexibility.
4. No Data Ownership
Most analytics tools save data at their side, which means that you lose data ownership and cannot guarantee your customers privacy.
Many companies are looking for alternatives to existing web analytics solutions because of these shortcomings. They want to explore new ways to collect streaming web data and deliver it to their data warehouse in real-time. Without compromising the data quality. Therefore, the biggest requirement for collecting and delivering web data to a data warehouse is flexibility. Websites change continuously, so the data collection should change simultaneously.
Start to End
Companies that are able to bring all their data together in a data warehouse can map all customer journeys from start to end. In earlier cases, we discussed churn reduction as one of those journeys. If your web data are available in real-time, you can immediately see and respond to customer drop-out, for instance by changing content. But you can also consider follow-up via e-mail. These actions can really reduce churn, as you make a real effort towards your customer. In the churn reduction case, the DimML data science language was mentioned as a way of connecting multiple data sources for a full view of the customer journey. It can also be used to stream data from website events to a data warehouse like Teradata or other data warehouses in real-time, from visits, campaign information and cpc to viewed product pages and order information. It is used for this cause at several enterprise companies in various markets, such as telecom, travel and insurance.
Cross the Edges of Data Silos
Of course, churn is just one of many possible customer journeys. But they all have one thing in common: you cannot understand and optimise them without a proper, reliable business-wide view that looks over the edges of existing data silos. And within that view, real-time availability of web data is essential. It is no longer an option to treat web analytics as a league of its own. If you are still analysing website interaction without the context of other data sources, you are not seeing the complete picture.
