Articles on data are typically about how to enhance, improve, or manage data accuracy. Data accuracy is important, it is critical, but it’s only one facet of the broader theme of data usability. The data you have might be 100% accurate, but it still might not be worth much at all if it isn’t usable.
At Iron Horse, we look at data and view it through the eyes of data usability (and decidedly not only data accuracy). Data usability, as we define it, is the level or the state of the data you have (typically evaluated at the field level) and how it fits its purpose. How do you know if your data is usable? Here are the five and a half pillars we look at when assessing a B2B SaaS organization’s marketing data for usability.
1. Accuracy.
In simplest terms, is the field based value correct or incorrect?
2. Precision.
For attributes that describe a company, opportunity, contact, etc. how granular and precise is the data? A good way to think about it is that for each attribute there’s a continuum from very broad to very precise that the attribute can be specified. As an example, think of collecting data for the field industry. At what level do your marketing and sales use cases require the data to be. Let’s say that a company is in manufacturing and builds boats; the continuum would look something like this (fromNAISCS):
- 31-33 – Manufacturing (broad)
- 336 – Transportation equipment manufacturing (precise)
- 3366 – Ship and boat building
- 336612 – Boat building (very precise)
- 3366 – Ship and boat building
- 336 – Transportation equipment manufacturing (precise)
3. Connectivity.
How integrated is the data with other data sources and data objects? As an example, having lead records that are accurate and precise is great, but if they are connected to other data objects such as accounts, support tickets, activities, web visits, product usage and campaigns then the data becomes more valuable. These data objects can then be used to drive additional personalization, segmentation and reporting options. In short, synergy occurs.
4. Population.
This looks at how consistently populated the field is across your data set. A field can be accurate, precise, and integrated with related information, but if it’s only found in a handful of records, its value decreases. For example, if a field is only populated five percent of the time it becomes difficult to use this field to drive personalization or any type of segmentation/filtering.
5. Standardization.
How structured is the data in the field? Structured data (think fields that are picklists or auto-complete) is critical for segmentation and filtering, but less important when it comes to fields primarily used for personalization.
5 1/2. Form.
Form is all about appearance and the way the data is expressed. We call this pillar 5 ½ as you can make a good argument that it really is part of standardization (so it really shouldn’t be called out separately). We agree with you, but it made for a catchy blog post title. Beyond that, though, we want to separate it out as data can be standardized, but at the same time have a form factor that makes it all but impossible to use. As an example, a former company I worked with had a CRM that was integrated with its ERP. Every night, the ERP force capitalized the first name, last name, and title fields. This made these fields nearly impossible to use for emails because an email that starts with “Dear JASON” or “Hi KELLY” and weaves a TITLE into the middle of a paragraph isn’t good marketing.
When assessing your marketing data, what else do you look at? Please tell me that you look at more than just data accuracy.
