Good Data: 4 Things You Can’t Do Without It In a Pricing Project

In today’s data-driven world, having good data is integral to the success of any pricing project. Good data underpins every aspect of your pricing operations, analytics, price calculations, and quote generation, defining the very essence of your business processes.  

How much businesses invest in their data maintenance tells a different story. The 2022 NewVantage Data and AI Executive Survey found that among leading organizations, many of which use data as a core part of their business, just 19.3% reported having an established data culture.  

Here at Pricefx, as a cloud-native pricing software company with over a decade of experience helping businesses with their most pressing pricing challenges, we find it surprising to observe so many businesses faltering in their data preparedness, despite its potential to bring multiple advantages to their pricing infrastructure.  

Many of the challenges our customers encounter in software implementation can be easily mitigated with a solid data readiness program, and we’re here to assist them in that journey so they’re set up for pricing success.  

In this article, we’ll explore 4 key ways businesses inadvertently sell their pricing software projects short as a result of bad data, and later show you what exactly good data looks like in practice.   

4 Things You Can’t Do Without Good Data for your Pricing  Project 

1. You Can’t Know Your Business Well Enough to Make the Right Pricing Decisions 

Simply put, it’s difficult to understand what’s really going on in your business if its data is incomplete or of a poor quality. When you don’t have a complete picture of your company’s past pricing decisions and their impact, you can’t accurately identify the right changes you need to make to your pricing moving forward.  

For example, if your company needs to assess its customers’ profit margins over time to identify overly-favored outliers or unusually high costs to serve, among other areas for optimization, you’ll quickly find that this task is impossible to carry out with inaccurate sales transactions data. Having poor data in your arsenal skews your analytics towards conclusions that don’t mirror the reality of your pricing history. 

2. You Can’t Roll Out Your Solution to All End Users  

Bad data brings a host of unforeseen technical roadblocks in testing that makes it difficult for product owners to have faith in the solution once it’s implemented.  If your project team encounters constant errors, inaccurate calculations, and skewed results at the User Acceptance Testing (UAT) stage due to poor data, this will undermine end users’ confidence in the solution and in effect slow down its adoption – even before it’s left the starting block. 

In short, your business can’t fully embrace a solution founded on bad data, as you can never be sure that what you’ll get from it is reliable and aligns well with your business goals.   

3. You Can’t Accelerate Time to Value 

Given the limited resources available for software adoption, we recognize that time to value is critical, and data readiness is an integral part of the equation. On the most basic level, your company wants to install a solution that’s going to help them – and you need to access that value as quickly as possible. Good data should be front of mind in the planning process because it is essential to how quickly your software project gets off the ground.  

In nearly all cases, something in the data gets missed that slows the project down. We recommend taking the time upfront to do a full inventory of all the data needed to support your operational processes, and once identified, ensure its quality and availability. This will help mitigate any costly delays or errors that may arise later on that can turn your 3-month implementation into a six-month, even year-long implementation. 

4. You Can’t Fit the Data to the Solution You Need in Scope and Scale 

The reality of most enterprise software out there is that they are founded on highly rigid data models, with a pre-set structure that leaves little room in its tables and fields for customization. Pricefx offers a more flexible platform to its users, which offers companies with specific business needs the configurability to build off a more custom data model that works best for those needs.   

With that in mind, it will be up to your business to ensure that its business data fits into the application’s view of a pricing solution.  

If you opt for a pricing solution based on a set data model, this will require reviewing and aligning your data with the predetermined field definitions in its master data tables. Solutions built on flexible data models won’t require this complex learning period, as they will mirror whatever the business needs to see in the system.

What Good Data Needs to Have for Your Pricing Project 

By understanding the key components going into good quality data, you’ll be well on your way to building a robust and effective data-driven solution and avoid roadblocks that could otherwise hinder the success of your pricing project. 

1. Your data is well defined 

A data element should be well defined so a user knows what kind of data they’re looking at. This includes what its format should be, whether it’s nullable or not, and other classifications.  

Once you establish a clear definition, the owner of this piece of data will then need to approve it. The errors you may come across in the testing stage may be as innocuous as the types of characters the system accepts or not, and having your data well defined at the beginning helps minimize these kinds of needless delays.  

2. Your data’s source is known 

Companies should know where their data lives if they want to retrieve and transfer it successfully. Its home could be an Excel spreadsheet, CRM software, or a cloud data management system, to name a few.  

Your company will need to know the answer to the key question, why is that data element housed there? While it’s usually clear how the data got there, knowing why it’s there helps determine whether it’s the right source or not. Doing so will smoothen an otherwise bumpy data transfer process.

3. Your data has a single source of truth 

Departments tend to house data in isolated silos, splintering the data storage system of the organization. This presents an implementation challenge for organizations seeking to make data-informed pricing decisions, as each location will contain duplicates or variations of the same data elements.  

In preparation for a move to a data-driven solution, a company should break down these silos and identify the authoritative source of truth for their data, for example merging its data into a central location or developing a canonical data model (a single data format for the company). You can already work with vendors at this stage, and as long as you can agree on the source of truth (which doesn’t necessarily mean all data needs to be in one place), your vendor can work with this.

4. Your data’s processes are known 

Another essential part of good data that is often overlooked is having a good understanding of the business rules and processes related to a data element, which speak to things like how new customers are entered into a system, how prices are set, how quotes are prepared, and other relevant factors. After all, your business processes will drive the data that’s needed, and finding that data is usually as simple as walking through each process and examining the data that’s used.  

No one knows the ins and outs of your business better than you, and this knowledge will ultimately serve as a guideline for how your data-driven solution will carry out its tasks.  

5. Your data’s owner is known  

Many companies tend to overlook who in their organization is accountable for a particular piece of data. When it comes time to prepare their data resources for a move to a data-driven solution, this lack of ownership over its data sources is often where companies get stuck. These individuals need to be identified early in the data readiness process so that the data is well defined and can be explained to external vendors when necessary. 

6. Your data is readily available and representative  

All data should be readily available – as in, available at the point in time the configuration team needs it. We recommend connecting with IT to make sure that the data can be extracted, either manually or automated, to avoid costly delays later on as a result of waiting for the data. 

Additionally, by providing representative data of all the user scenarios upfront, companies can avoid the frustrating rework required after they inevitably introduce newer, more representative data sets to their software provider, who may be already halfway into configuring a user story. If, for example, you classify your customers in seven ways and you provide data for five of them, it will be difficult to predict with confidence that those five will be representative of all seven.  

7. Your data is complete   

In an ideal scenario, data is not missing any key fields or gaps which would cause orphaned records, data whose “parent” does not exist. As mentioned earlier, your data also needs to be representative of all the conditions you set prior to user acceptance testing.  

Importance of Collaboration in Data Governance 

We appreciate that data governance is a large topic. There is no universal way an organization can reference to diagnose and approach its data health – this is highly dependent on its business processes and goals, and how well its data has been maintained and focused on in the organization.  

With that being said, facilitating productive conversations between the key stakeholders of your software project is an important first step to promote strong data governance practices.  

IT and business users often have different perspectives on software functionality – IT brings technical understanding while business knows how to use the tool to bring value. It’s important to align these perspectives early in the project to ensure successful implementation within the planned timelines and budget.  

Because while business may think the project is ready to hit the ground running, IT may have a different timeline in mind depending on the state of your data. Getting these two groups together at the beginning – and more importantly, getting them to speak the same language, which can be a project in and of itself – will cut down the extra lead time needed to set your data up for success.

Final Thoughts: Key Steps to Prepare and Organize Your Data for Successful Project Execution 

Now that we’ve got you thinking about what good data looks like and what you can’t do without good data in place, we’d like to close out with a few next steps to keep in mind when kicking off your data readiness process: 

• Start preparing data well before the start of your project by identifying the data sources, their owners and related processes. 
• Identify key dependencies such as IT time required to prepare data, and other initiatives that compete for resources or that must be coordinated. 
• Identify key team members and subject matter experts associated with the data. 
• Treat data readiness as a project in and of itself. 

To tie this all together, your organization will first need to get a full picture of its data, identify the scope of the problems you set out to solve with the tool, and be willing to undertake the effort.  

Ready to take the first step in your data readiness journey? Follow our guideline to get your data ready for a successful pricing software implementation: