#ArtztOnData: Understanding Data Orchestration

Data Orchestration solutions solve some of the most challenging issues affecting companies that want to increase the value of their data assets. Given its centrality to emerging data strategies, it’s worth taking a moment to set down a clear definition for Data Orchestration.  

A Brief Definition of Data Orchestration

What is Data Orchestration? At a high level, it’s a process, borne of two insights. The first insight arose among people who manage data in sales and marketing operations. They realized that most, if not all of their data sets, contained duplicative, low-quality data, affecting their ability to succeed in their work.

The second insight came from the search for a solution to the first problem: How can you make bad data better? The answer was certainly not to be found in manual processes. Manual de-duplication and data enrichment, to name two out of many steps in data quality management, are far too slow and error-prone to scale. Data Orchestration is the solution.

Data Orchestration automates the data quality improvement process. The automation involves orchestrating the actions of various systems that are required to process steps. For example, data enrichment invariably means reaching out, system-to-system, to a data enrichment provider. This process must be orchestrated in order for it to work on an automated basis. In this way, Data Orchestration is analogous to business process orchestration (BPO), in which systems send and receive procedure calls to execute a multi-step business process.

Solving the Data Quality Problem

Low quality data presents itself in the form of duplicative, incomplete, inaccurate and non-normalized data sets. Let’s break these down one at a time:

–         Duplicative—Given that sales and marketing data generally flow into an organization from multiple points of input, such as web forms, list uploads, and APIs, it’s inevitable that contact databases will contain multiple records for the same person. Indeed, if your marketing campaigns are working, they will likely snare the same leads over and over. That’s a good sign. It means prospects are engaging with your brand. However, if you don’t clean up those duplicates, you’re going to start making embarrassing mistakes like sending five postcards to John Smith at Acme, Inc. in Peoria, Illinois.

A data orchestration solution will activate its de-duplication workflow when a new data record is generated. The workflow automatically goes through the steps of looking up duplicates, comparing possible duplicate matches and determining the best way to merge them into single records. The solution orchestrates the sequence of lookup, comparison and merging.

–         Incomplete—Data intake processes frequently lead to incomplete data records. One reason for this is because you can’t make every field on a web form mandatory. That will reduce form submissions. You end up with records that are missing data elements, e.g. J S at Acme. 

Data orchestration will handle the process of enriching incomplete data. For example, to turn “J S” into “John Smith,” a data orchestration solution may look up this person’s information on external databases such as LinkedIn. Similarly, the solution can append 5-digit zip codes so all records contain full 9-digit zip codes. It can add SIC codes and industry categories to company records and so forth.

–         Inaccurate—Prospect and customer lists contain inaccuracies. These come from data input errors, such as when someone takes a name over the phone and keys it in wrong. Errors can also come from acquiring lists that come with mistakes baked in, e.g. Jon Smyth at Acmeinc. Data orchestration can compare data records with external, presumably correct data sources and fix mistakes in the database—so Jon Smyth’s entry is corrected to John Smith.

–         Non-normalized—Normalization refers to ensuring that each field in a database follows a standardized, uniform “normal form.” For example, if the state of Illinois is represented by IL, Illin., Il, and Illinois in a customer database, those variations will cause problems for effective use of that data. Data analytics, for instance, becomes less useful if business intelligence (BI) tools cannot accurately group contacts by state. Data orchestration can impose a normalization pattern onto the data set, e.g. that all references to Illin., Il, and Illinois should be replaced by IL.

Data routing is a further quality problem, though it’s not directly about the content of the data itself. Rather, if data is not routed to the right people within an organization, it will not have the desired impact. Consider the following situation, which is unfortunately quite common: John Smith from Acme, Inc. fills out a web form so he can download a white paper. The problem is that he’s already in the prospect database, with his account being handled by an account team.

If data routing is not set up properly (or at all), the account team may not learn that he’s downloaded the white paper. They’ll miss this sign of interest and engagement. Worse, a separate account team might reach out to him, wasting time, causing confusion and creating a poor impression. Data orchestration can create a known, repeatable process for the correct routing of data.

Data Orchestration Outcomes

As data orchestration goes to work on customer and prospect data, it renders that data more valuable as an asset. The increase in value manifests primarily as the result of accelerated sales and marketing cycles. Sales teams can work more productively because they are not wasting time on dead end leads and calling prospects who have already been contacted, and so forth. These seemingly minor gains add up. The result is the potential for faster revenue and earnings growth, the drivers of entity valuation.

More accurate, enriched and normalized data facilitates effective marketing campaigns. Smarter targeting leads to higher rates of response and engagement, which in turn drives sales growth. Better tuned marketing also drives strong brand engagement. 

Data orchestration also enables more effective data analytics, which generally leads to better executive decision making. Armed with accurate data visualization and reporting, business managers can be well informed as they set strategic direction and make investment decisions. Data orchestration allows everyone to work smarter.

Conclusion

Data orchestration is gaining traction with marketing and sales organizations. They see it as a solution to long standing difficulties they have experienced with data quality management. By applying data orchestration to customer and prospect data, companies can speed up their revenue operations and grow with less friction holding them back. 

Learn more about RingLead's approach to data orchestration: https://www.ringlead.com/