Activate Data Natively In Tools
A lot of our tips have focused on creating data infrastructure so we can pull out and combine data from different sources, and so it might be surprising to see that tip number five is then putting that data back, allow us to explain…
What is Data Activation?
Typically a data model will be surfaced in reports and dashboards. A person then needs to understand and interpret that data, and make an action from it. While this sounds straightforward, it presents a number of challenges.
Data literacy: In terms of data literacy, there may be a bigger problem with basic numeracy skills then you might imagine. Every few years the OECD runs a large study of 245k people across 38 countries, examining the basic numeracy skills of adults. A 2018 survey found that England was ranked 21st and America 28th out of 38 countries. A shocking 6 in 10 Americans struggle to “recognise and work with mathematical relationships, patterns, and proportions expressed in verbal or numerical form; and can’t interpret and perform basic analyses of data and statistics in texts, tables and graphs.”
Data Trust: For those who can interpret graphs and charts, there is a huge issue around trust in the data itself. An Experian study polled C-suite managers and found that whilst 98% said that having high-quality data was either “extremely important” or “important” in achieving their business objectives, 40% said they didn’t trust the insights they get from their data and 28% described their current customer/prospect data as inaccurate. Where people do not trust in data, they are more likely to ignore what the metrics are telling them.
Bias: On a similar theme, a complaint we often hear about C-suite managers is that when presented with the correct data, it is ignored in favour of ‘experience’ or personal bias. Personal bias significantly impedes the ability to act effectively on data and insights, as it distorts perception and interpretation of information. When decision-makers allow their subjective views, preferences, or experiences to influence their analysis, they may selectively focus on data that confirms their preconceptions while ignoring or downplaying evidence that contradicts them.
Human Error: Manual data activation might entail tasks like; downloading a list of event attendees and uploading it to a marketing tool like HubSpot, or cross-checking data within visualisation tools before updating it within their marketing platform. The cons of this approach are obvious; manual data activation is a time-intensive process, relying on human input introduces the risk of errors, and manual processes can’t scale efficiently, as the number of data points to manage becomes overwhelming.Where you have made significant time and effort to automate your data modelling, it does not really make sense to then serve this up for a person to activate manually.
Within Existing Tools: Many companies opt to use their existing tools, such as Customer Relationship Management (CRM) and marketing tools like HubSpot or Salesforce, for data activation. Leveraging existing tools is convenient, as employees are already familiar with their functionalities. It allows for a quick start in data activation without the need to invest in new systems and so long as companies consider data quality and oversight of these tools, then they can act as that single source of truth. But therein lies the problem. Relying on these tools can magnify data quality issues. Even something as simple as a misclassification of a sales lead could mean substantial time and marketing spend wastage. While these tools often incorporate features like propensity models and lead scoring, it is quite common for them to employ simple, one-size-fits-all models, providing generic insights that may not be specific to a company’s unique needs.
So What Is Reverse ETL?
When it comes to activating data, there is another route called Reverse ETL. In a normal ETL process data is extracted from its source, loaded into a data warehouse, transformed and combined through modelling into business metrics and then visualised. What reverse ETL does is it takes that modelled data and pipes it into any application. So you might use 4 different data sources to understand which customers are going to churn. Once you have that data modelled, instead of it living in a reporting tool, you can send it directly to a CRM for your team to intercept. This approach has a number of advantages:
Centralised Source Of Truth: Reverse ETL relies on data from within your data warehouse/lakehouse which means it doesn’t create another data silo and ensures that your warehouse is a single source of the truth. This would differ from many CDPs which are used by marketing teams, which can create a silo in which they are relying on one tool for data, and the rest of the organisation uses another. Wherever there’s two different data tools being used- there’s always a gap or a difference in the numbers! With reverse ETL your entire organisation is completely aligned with the same metrics. Maintaining a single source of truth in the data warehouse while ensuring that operational systems receive only the necessary and relevant data. This improves data quality and compliance with data regulations.
A Guiding Light: Your modelled data can be used to create lists, flags and notifications inside the applications your teams use and are familiar with, making data activation easier and more effective.In this way your data will act as a guiding light for your experienced team to then make a decision. This is very useful in that it lowers any data literacy or bias issues, as the model will suggest what actions need to be taken.
No Manual Errors: Automating the flow of data between systems reduces manual data entry and the potential for errors. This streamlines operations, saving time and resources while ensuring data consistency across platforms.
Data Democratisation: Reverse ETL allows non-technical users to access and use data without requiring SQL knowledge or access to the data warehouse. This democratises data across the organisation, empowering more teams to make data-driven decisions. No data solution is successful if people aren’t using it and at the most basic level, by activating your data in tools teams are already comfortable with, it means they don’t need to learn a new interface.
Scalability:Reverse ETL processes can be scaled to handle increasing volumes of data and more complex data flows. This makes it suitable for growing businesses that need to maintain efficient data operations as they expand.
At What Stage Should You Consider Reverse ETL?
Once your data is centralised in a data warehouse or data lake, the next step is often to make this data actionable. You should consider using Reverse ETL right at the start of your idea generation phase. Why? Because it is always good to think about activation first and foremost. How will your business stakeholders interact with and use your data? What are the tools they are already using today? How can you make mass decision-making easier?
There are of course a few other considerations:
Data Readiness: Ensure your data is clean, well-structured, and in a format that can be easily extracted and loaded into target systems.
Integration Complexity: Assess the complexity of integrating with various operational tools and whether the Reverse ETL tool supports these integrations.
Scalability: Choose a Reverse ETL solution that can scale with your data needs and business growth.
Compliance and Security: Ensure that data handling complies with relevant regulations (e.g., GDPR, CCPA) and that robust security measures are in place to protect sensitive data during transfers.
Conclusion
Activating your data in tools that teams already use helps to ensure timely, relevant decisions are made without the manual manipulation that can lead to human error. It helps bridge that gap between technical and non-technical users. If you are curious about implementing this and want to see if your organisation is ready, be sure to get in touch with the friendly team here at 173tech.