Learning Library

← Back to Library

Four Pillars of Data Quality

Key Points

  • Poor data quality can undermine business outcomes just as low‑quality ingredients ruin a chef’s dishes, damaging a company’s reputation.
  • Accuracy means data must reflect reality; unfiltered bot traffic can skew lead‑generation metrics and produce inaccurate results.
  • Completeness requires all necessary fields (e.g., names, emails) to be filled, otherwise the data set provides an incomplete customer picture.
  • Consistency and uniqueness ensure uniform formatting across sources and eliminate duplicates, preventing mismatched records and inflated lead counts.

Full Transcript

# Four Pillars of Data Quality **Source:** [https://www.youtube.com/watch?v=5HcDJ8e9NwY](https://www.youtube.com/watch?v=5HcDJ8e9NwY) **Duration:** 00:03:49 ## Summary - Poor data quality can undermine business outcomes just as low‑quality ingredients ruin a chef’s dishes, damaging a company’s reputation. - Accuracy means data must reflect reality; unfiltered bot traffic can skew lead‑generation metrics and produce inaccurate results. - Completeness requires all necessary fields (e.g., names, emails) to be filled, otherwise the data set provides an incomplete customer picture. - Consistency and uniqueness ensure uniform formatting across sources and eliminate duplicates, preventing mismatched records and inflated lead counts. ## Sections - [00:00:00](https://www.youtube.com/watch?v=5HcDJ8e9NwY&t=0s) **Chef Analogy Highlights Data Quality** - The speaker explains how poor data quality—measured by accuracy, completeness, consistency, and uniqueness—can damage a business’s results, using a chef‑restaurant metaphor and lead‑generation examples. ## Full Transcript
0:00your company generates lots of data but 0:03the business outcomes you gain from that 0:05data can be largely affected by data 0:08quality 0:09to use an analogy imagine you're a chef 0:12and you have the highest accolades in 0:14the industry a highly experienced team 0:16but when the ingredients come in those 0:18are poor quality ingredients picture 0:20rotten tomatoes rotten onions so when 0:23you go and make those entrees the end 0:25result is poor quality and your 0:27restaurant's reputation suffers 0:29this is the same impact that poor data 0:31quality can have on your business 0:33causing your company's reputation to 0:35suffer as a result 0:37there are a lot of different factors 0:38that can impact data quality such as the 0:40number of sources or the size of your 0:42company but today i want to talk about 0:44four main qualities within data itself 0:47accuracy unit completeness consistency 0:51and uniqueness 0:52and i'm going to talk about them through 0:53the lens of a lead generation company 0:57starting with accuracy 0:59accuracy is about the current state of 1:01your data versus reality 1:04so for my lead generation company 1:06imagine i'm driving traffic to a website 1:10and all of a sudden i get a sudden spike 1:13in usage from bots that hit the click 1:16generation 1:17if i don't account for this spike when i 1:19go and pull that data at the end of the 1:20day it's not going to reflect reality so 1:23it's not going to be accurate 1:25next i want to talk about completeness 1:28which is about how you have filled out 1:31all the required fields 1:33in your data set 1:34so let's say i'm launching a survey 1:37campaign 1:39and i'm collecting 1:40names and email addresses 1:42but i don't require this field so when i 1:45go and pull that data i notice that some 1:47of my participants didn't put their name 1:49some of my participants didn't put their 1:51email so when i go and pull that picture 1:53of the client of the customer i have an 1:55incomplete data set and a complete 1:57picture 1:59next we talk about consistency which is 2:02about how uniform 2:04your data set is throughout different 2:06data sources 2:08so back to my lead generation example 2:11let's say i'm driving traffic for a drop 2:14shipping campaign 2:15and i have my procurement team 2:17collecting zip codes and my marketing 2:20team collecting zip codes but my 2:22procurement team is looking at them 2:25in a five digit format while my 2:27marketing team is collecting them in a 2:29nine digit format 2:32when i go tap into both of these 2:34databases and pull the customer profile 2:36it might be incomplete because those zip 2:38codes don't match up throughout my 2:40systems 2:42and lastly there's uniqueness which is 2:45largely tied to the number of duplicates 2:49i have in a data set 2:53so in my lead generation context you can 2:56imagine having 50 000 leads at the end 3:00of the year but when i actually go into 3:02those leads i realize that 20 3:05are duplicates from customers who filled 3:08out the information previously so now 3:10when i go and pull that report i 3:11actually have 20 percent less data and a 3:14lot 3:15less 3:16positive looking picture for my company 3:19so 3:20looking at these aspects it's easy to 3:22think wow there's a lot of manual 3:23inspection here how can i go through all 3:25of my data and understand these 3:27resources these 3:29qualities right well you can actually 3:31leverage machine learning and ai to 3:33automatically sense these key features 3:35as data enters your system saving you 3:37time and manual inspection 3:40if you're curious about these ai 3:41features check out the links below and 3:43if you're curious about technology 3:44subscribe to the channel thank you