Why Data Quality matters and what is a clean database?
Imagine Dirk, a construction worker. He has his set of tools which are indispensable to do his work. Dirk organizes his tools by compartment, going from hammers to screwdrivers, drills, etc. By doing that he knows where to find what tool when he needs it and work more effective.
Dirk's idea isn’t about finding his tools, it’s also about saving time and using that time to deliver a great service. When you apply Dirk's case to your company you will see that having a clean database is crucial for your business to run smoothly. Here is where intelligent words such as data governance pop up.
And there's a link with content. Your staff will have more trust in the content they produce, be more productive while creating content, signal potential opportunities, etc. Because your data is clean, your marketing department can then target more accurately which directly affects the ROI of your running campaigns.
Here's the good news...there's a straight way forward improving your data quality...Your data quality checklist. More on that below.
First, Do you want to locate the problem(s) you are having with your data quality? Do you want to know how qualitative your data is? The checklist below can help you identify if your problem is coming from the quality of your data.
How to tell you have data quality issues? You'll know when:
- You have to manually enter data into your database;
- Your contacts do not receive your message;
- You have a high bounce rate
- You notice a decrease in the effectiveness of your campaigns
- Your ISP-trust score is decreasing over time;
- You notice that your data mostly consists of duplicate data;
- You’re struggling to convert leads into customers;
- You notice a decrease in lead-activity.
One or more of the cases above sound familiar to your situation? If yes, take the next three steps:
- Centralize your data
- CAVCAT your data
- Activate your data
Step 1: Centralize your data
Imagine that Dirk is looking for a new drill online. He ends up on your website and watches a demo from one of your drills, read its reviews, downloads a brochure, etc. Because Rob has access to a lot of information, he doesn’t hesitate to leave his details on the contact page to show that he is interested in your drill.
The cookies you add on your website track Rob’s movement, even if he closes and re-opens your website. For example: the pages he viewed, brochures he downloaded, video’s he watched, etc. are all tracked and stored into your cookie. But what do cookies do?
Cookies send all the tracked & stored information directly to your database. Creating and updating leads automatically in your database. Giving your business all the necessary information & tools needed to convert Rob from a lead to a customer with relevant, personalized content.
Step 2: CAVCAT your data quality
Now your database is connected to your cookie and automatically being updated. It is time for consistency in your database. No more endless rows of unorganized, unstructured data. You and your staff want to find what you are looking for in a matter of seconds and use it.
The process of creating consistency by organizing and structuring your database is often referred to as ‘The normalization process’. You define the properties of each data field using ruling types. We call it the CAVCAT-process:
- Completeness of the data fields: are the most crucial data fields (names, company, mail, ..) complete? Does the mail address field contain an “@”?
- Accuracy: How well does the data reflect the lead? Does the data he entered match his behavior on your marketing assets? Is the data a match with what your company is looking for (CEO, CMO, CFO, Decision Maker)? The more you automate your Marketing processes, the more accuracy becomes important. When you don’t take accuracy into account you will only amplify the ‘Garbage in - Garbage out’-principle, which is something we all want to prevent from happening.
- Validity: Does the data conform with the rules set for that data field? does the phone number field consist of numbers?, etc.
- Consistency: are the patterns over all the data fields unified? Is birth date everywhere MM/DD/YYYY or does it differ?, Is country location for the phone field “+3X” or “003X”, etc.
- Availability: Is the data unique in the database? Are there duplicates on the “mail address” or “full name” data field?, etc.
- Timeless: What is the impact of time on the value of the data? Does the data rely on information spread over a period of time in order to be accurate or not?
When you’re ready implementing CAVCAT on your data, you’re all set to start optimizing your database for an even smoother and more productive way of working.
Step 3: Activate your data
Just like Dirk's toolbox, we have re-arranged all our data fields. Now in order to be as efficient as Dirk, we should order our data per compartment. Or simply segment our different data fields into a larger group. So that our staff can easily correlate and navigate to larger groups to increase staff productivity.
Some practical examples are:
#1. Segmenting on the company name: Imagine that Dirk and his boss, Marc, are both interested in drills and subscribed to your newsletter. However, both of them entered the same company name in a different way. Dirk entering “We Construct 4 You” and Marc entering “WC4Y”. In order for your database to represent Dirk and Marc under the same company, you must define that both of the companies entered values are listed under the same group so that your staff can easily see that they both belong to the same company.
#2. Segmenting on function: whenever you market to your leads, your communication could differ based on the function of the lead. For example, when you would market to Dirk, your communication could all be about short-term actions like “20% off of your drill” and communications to his boss, Marc could all be about long-term investments like partnerships. This makes it easier for your staff to use the right tone-of-voice to each and every segment.
#3. Segmenting on geographical area: when you create a segment based on geographical area, typically connected to the ZIP/Postal code, you create a better overview on well the area is staffed. Giving your company the ability to increase and decrease the amount of sales-people in a certain district or plan the most time-efficient routes for your staff in advance.
Using segmenting criteria definitely comes in handy when you’re working with a large database. This way you learn more from your database and you are able to make better analysis’s to use for your future marketing strategies.
Make your clean data shine
To get the most out of your clean data, you should act on it wisely. You have the tools, now it’s time to get the ball rolling. Your data is the gate to giving your customer the exceptional experience they deserve at the perfect moment.
Firstly, the cookies that are connected to your database can also be used for progressive profiling (more on that in the blogpost about the functionalities of Marketing Automation). Meaning that each time that Dirk fills out a form on your website, he will be asked for different details. A first form could ask for name and email address, but a second could ask for company and phone number. This way you collect more, relevant information to tailor each marketing message to each and every lead. Consider that whaty they tell you about them selves is always less important than the behaviour. What they do (what they read for example) is much more important and clarifies much more the intention of their engagement with you.
Secondly, it would be a shame if all your previous cleaning would be outdone by time. Because yes, just like we all do, data ages too. People switch addresses, develop interests, build relationships, etc. Therefore, it is wise to frequently use CAVCAT on your database. Depending on your field of activity. For example, a bookstore might want to check his data on a 3-month basis when a car dealership might want to check on an annual basis. 3 years ago database eroded at 10% per year.
Nowadays that's 30%. 30% of all your data erodes every year. Now if we know on average only 1% of customer data is ever used those are scary statistics...
In the end, you can define highly productive and effective strategies based on the different segmentations you create. Making all of your business-efforts return the expected results
Want to find out more? Read Obsessed so you can start decoding your data landscape and reboot your revenue engine.