Before buying new equipment, a company typically looks at the cost benefit analysis, ROI, operational efficiency and the required time needed to operate the new equipment. But when an organization acquires data (regardless of the source) the majority look only at its original cost.
That’s a shame. Poor data quality can have a negative impact on your lead management system, the number of hours required for sales and marketing spends on menial tasks, and can lead to additional costs to increase the quality of the original data. So what is the true cost of the bad quality of data?
To answer the question one must define data quality. Data quality can be defined by 4 C’s: complete, current, correct and consistent.
Let’s look at an example of a record which is incomplete, incorrect, not current and not normalized (consistent). Say you acquired the below record, which has an original cost of $1.00. Most marketing departments will allocate this cost to the data acquired and stop there. However, consider the following costs associated with each area of data quality.
|Michael Smith||US||ABC Company||150Mfirstname.lastname@example.org||123-4556||Marketing Manager|
Incomplete data: In the above example the address, industry, and area code are all missing. For efficiency purposes you append third-party data which costs $1 per record. The two to three days you spend waiting for your data to be improved, you’re losing valuable time you could spend following up with a lead that’s still hot. You are also giving competitors ample opportunity to contact your lead.
Incorrect data: Now that you have a complete record, you send an email making an offer. You get a bounceback. This impacts the business in two crucial ways. First, your email deliverability score decreases. (For more on this subject check out the Grande Guide to Email Deliverability & Privacy.) Secondly, you are left with a possible lead without a correct email address. Perhaps you pass it to the sales guy to call. Assuming he reaches Michael Smith, he probably spends an extra 10 minutes explaining to Michael what the bounced email said before beginning his usual ‘sales pitch’. If the sales person earns $20 an hour, that’s another $3 to add to the price of the record.
Not Current Data: While the sales person is on the phone he realizes that Michael Smith is no longer the marketing manager but has been promoted to Marketing Director, and that he should be really talking to the new marketing manager. So now the sales person has to add a new contact to the database, which takes another 10 minutes. He also phones the new contact, which takes another 10 minutes and adds another $6 for this new record.
Not Consistent Data: Now that the record in the database is correct, current and complete, the marketing manager decides to do a direct marketing campaign to this record. However, he realizes that the address is not standardized. He sends it to a third-party that charges $0.50 to standardize the address.
If you add all of the costs together you have the following cost for this record:
|Original Cost||$ 1.00|
|3rd Party Appending||$ 1.00|
|Address Standardization||$ 0.50|
|Correcting Information||$ 3.00|
|Updating Current Information||$ 6.00|
|Overall Cost||$ 11.50|
The difference between the ‘booked cost’ of $1 to the actual cost of $11.50 is $10.50 per record (multiply that by 1,000-10,000 records per campaign), with extra two to three days of waiting prior to the following up on a lead and the sales person wasting at least 30 minutes of his time. A marketing manager in this situation might look at the result of her campaign and think this data acquisition initiative was successful. But she’s looking at the original cost without considering the quality implications. This will lead to bad decision making in the future, and ultimately a decrease in revenue.
While high quality data which is current, correct, consistent and complete might come with a higher sticker price, it’s actually cheaper when compared to the true cost of bad data.