- The term single source of truth is constantly repeated in IT, particularly with regards to data.
- We cover the accuracy of this term.
Terms are used in IT that are sometimes valid and sometimes are inaccurate marketing terms. The term single source of truth is sometimes used to explain being able to find nonconflicting information from one place.
Our References for This Article
If you want to see our references for this article and other related Brightwork articles, see this link.
Lack of Financial Bias Notice: The vast majority of content available on the Internet about SAP is marketing fiddle-faddle published by SAP, SAP partners, or media entities paid by SAP to run their marketing on the media website. Each one of these entities tries to hide its financial bias from readers. The article below is very different.
- First, it is published by a research entity.
- Second, no one paid for this article to be written, and it is not pretending to inform you while being rigged to sell you software or consulting services. Unlike nearly every other article you will find from Google on this topic, it has had no input from any company's marketing or sales department.
How Valid is The Term Single Source of Truth
The term single source of truth is defined by Wikipedia as follows.
In information systems design and theory, single source of truth (SSOT) is the practice of structuring information models and associated data schema such that every data element is mastered (or edited) in only one place. Any possible linkages to this data element (possibly in other areas of the relational schema or even in distant federated databases) are by reference only. Because all other locations of the data just refer back to the primary “source of truth” location, updates to the data element in the primary location propagate to the entire system without the possibility of a duplicate value somewhere being forgotten.
This definition is more technical than most people understand when they use the term. Normally people use the term to mean that the source can come from one system.
Wikipedia goes on to say.
The “ideal” implementation of SSOT as described above is rarely possible in most enterprises. This is because many organisations have multiple information systems, each of which needs access to data relating to the same entities (e.g., customer). Often these systems are purchased “off-the-shelf” from vendors and cannot be modified in non-trivial ways. Each of these various systems therefore needs to store its own version of common data or entities, and therefore each system must retain its own copy of a record (hence immediately violating the SSOT approach defined above). For example, an ERP (enterprise resource planning) system (such as SAP or Oracle e-Business Suite) may store a customer record; the CRM (customer relationship management) system also needs a copy of the customer record (or part of it) and the warehouse dispatch system might also need a copy of some or all of the customer data (e.g., shipping address). In cases where vendors do not support such modifications, it is not always possible to replace these records with pointers to the SSOT.
Yes, this is correct. Data is stored in multiple locations — and so the term “single source” is not really a reasonable expectation.
How Software Vendors Push the Single Source of Truth Idea
Observe the following screenshot from the software vendor Atscale.
This is a typical presentation from a vendor on how companies can benefit from SSOT if they purchase their product. Having “everything in one place” is a common promise of software vendors, but no matter how many times the promise is made, data is still distributed across different systems. Atscale also repeats meaningless statements about “common business language.”
Oversimplifying the Goal of SSOT
However, vendors and consulting firms frequently oversimplify how easy it is to arrive at a single truth source or how feasible this objective is. This is playing with companies’ desires because companies are normally frustrated about duplicating data across systems, so vendors and consulting firms frequently promise that a single source of truth is possible.
Observe this quote from Mulesoft.
A simple example of this is search engines. Let’s say a searcher enters a query about their favorite restaurant. Google aims to be the single source of truth for anything they may need to know about this restaurant; such as, the restaurant’s hours, phone number, local locations, menu link, ratings, and popular times. Google is bringing data from many sources (Google maps, the restaurant’s website, Yelp or Google ratings, etc.) to be the searcher’s single source of truth for the data they may want to know about that restaurant.
The problem is that while Google may want to be something, it isn’t. Google might be the most prominent source of information on a topic, but it is not the single source of information.
Mulesoft then promises companies that an SSOT is a feasible goal.
Creating a single source of truth ensures that businesses are operating based on standardized, relevant data across the organization. Without a single source of truth, data sets exist in siloes and each department operates as a black box. Implementing a single source of truth enables business leaders to make data-driven decisions based on the data from the business as a whole, rather than from compartmental data silos.
How Does Mulesoft Recommend Obtaining an SSOT?
For a business to obtain a SSOT, all data components from the various systems across the enterprise will aggregate their datasets to the primary single source of truth location.
This is a data warehouse. However, I have worked for many clients that have had a data warehouse. In fact, over several decades, I don’t know if I ever worked for a client that did not have a data warehouse. But I also don’t recall any of these companies ever having a single source of truth. So clearly, a data warehouse does not result in an SSOT.
More Overpromising of an SSOT
The following presentation of the importance of an SSOT is provided by an article by Brent Dikes, who paid for his article Single Version Of Truth: Why Your Company Must Speak The Same Data Language, to appear in Forbes. (Forbes no longer has any writers on staff and rents out its website for anyone who will pay them).
On average, marketing departments are using 12+ marketing systems. With the high volume of digital marketing tools that are available, it’s easy for marketing groups to exceed this average. When each of these tools is spitting out different numbers—even similar systems such as Google Analytics and Adobe Analytics—it can be hard for marketers to know which numbers to rely on. Metrics that share a similar name may be measuring very different things. Often differences can come down to when, where and how the data is collected as well as the settings and processing within the separate tools. Without a single version of the truth, it’s easy for marketing teams and entire companies to end up working feverishly on the wrong activities that fail to move the needle.
This is a prevalent approach that is used to mislead IT buyers.
It states the problem. However, it does not necessarily follow that just because there is a problem that there is a straightforward solution.
The SSOT is essentially a utopian vision that leaves out how difficult it is to attain. By oversimplifying the solution, any presenter can make it appear as if they have the solution.
Marketing and Sales Will Agree on What Constitutes a Lead?
Observe the problem is apparent from just playing forward the implication of another quote from this same article.
Your marketing automation system will apply filters and remove duplicates to create a lead that has been “qualified” by marketing. When these marketing leads are passed over to your sales team, they may only accept some of them as legitimate sales opportunities they want to pursue. If you want to optimize your business for “leads”, which variation do you choose? Any misalignment across the business can have serious consequences as teams end up pulling in different directions—many of which will be the wrong ones.
So what is the SSOT in this example?
Is it the marketing leads or the sales leads that the sales team considered legitimate sales opportunities? Marketing and sales generally do not see things the same way.
Marketing tends to want to count all leads they hand over to sales. But sales only care about legitimate leads, not just leads. Marketing has one set of criteria for what counts as a lead, and sales had another. But this is an important example because it demonstrates that there is not one single agreed-upon number. And this does not have anything to do with what system might be used.
The Lack of a Single Source of Truth Within Just One System (The CRM System)
CRM systems are notorious for having inferior quality data in them.
Reason #1: Different Incentives of the Uses of the CRM System
Sales directors are generally much keener on having a CRM system updated with leads than salespeople are interested in placing their leads into the CRM system.
Reason #2: Different Incentives of the Uses of the CRM System
- Sales directors use the CRM system as a type of surveillance on salespeople.
- Sales directors tend to micromanage salespeople and use the CRM system to do it. Many sales directors obtain their positions for political reasons and don’t have much sales insight. Therefore many salespeople don’t actually want input from their sales directors.
- Companies will take a salesperson’s leads when they are fired, as they see them as theirs, even if the salesperson brought those leads with them from a previous employer.
Reason #3: The General Incentives and Orientation of Sales
Both sales directors and salespeople often exaggerate the likelihood of lead buying so that they can keep their jobs. So again — what is the SSOT? Is it the likelihood of leads converting entered in the salesperson’s CRM system by the sales director or by an objective third party with no quota attached to their compensation.
However, right after the case has been made that leads in the article, the interpretation is subjective. Dikes goes ahead and states that one needs an SSOT.
When you introduce a single version of the truth, the central issue isn’t accuracy—it’s about alignment and buy-in.Jeanne Ross at the MIT Sloan School’s Center for IS Research stated, ”Getting to one version of the truth doesn’t have anything to do with accuracy, it has everything to do with declaring it.”
The central issue is not accuracy? How can the central issue with data not be accurate? That does not make any sense. But even if we throw that comment out, what is the right SSOT?
Now notice another flaw in the logic being presented.
It’s better to move ahead with good data that can be improved over time rather than waiting (and waiting) for perfect numbers.
That point is debatable, as it depends upon how high the stakes are and how “good” the data is.
But secondly, this is not related to the point of the SSOT for the leads, the number provided by marketing, or that accepted by sales.
I hate to be the one to break this to Dikes and Ross, but marketing and sales will never agree as to what the SSOT for leads is. Marketing has a financial incentive to count the leads as higher in potential (thus justifying their existence). Sales will have a financial incentive to count the leads as lower in potential, as they actually have to close the leads.
Today, the predominant challenge for most businesses isn’t a lack of data—but choosing which numbers to focus on. When most companies have multiple data systems and a wealth of data to draw from, it’s easy to feel data-driven but not actually be data-driven. Establishing a single version of truth is about separating the strategic signal from the operating noise of your business.
Saying it does not make it so.
If executives get into a meeting and develop a consensus on what the numbers should be, and everyone in the company agrees, that also does not make it so.
Adding The Term Data-Driven to SSOT
And using the term “data-driven” is just another buzzword that is now being applied to the term SSOT. It does not enhance the actual ability to know what are the correct result. Recall that oftentimes things being determined are a forecast. Well, a forecast is not just data. It is data combined with a particular forecast method. And different methods applied to the same data result in different forecasts.
Therefore saying that one needs a “single source of data” is a bit like saying one needs a “single forecast method.”
Everyone at your company must speak the same data language—aligned metrics, dimensions and definitions—to be truly successful with data.
Well, good luck with that. But secondly, how can individuals from different departments with different financial incentives all agree. This is not a “data language.” This is agreeing on what the numbers should be that are agreed upon. These companies are relying on a data language. It’s called the Arabic number system. The issue agrees on what the numbers need to be, what the forecast should be, etc.
I consider the term SSOT a meaningless marketing term. There is never a single source of truth because there is never just one system. When the term is used in articles, the logic of SSOT is easily contradicted.
- The term is basically not used by technical people but is frequently used by software salespeople and marketing types.
- Promising a single source of truth is simply a way to sell software and or consulting services.
- Vendors and consulting firms have been promising an SSOT for decades, and yet the SSOT remains elusive.