ellipsis flag icon-blogicon-check icon-comments icon-email icon-error icon-facebook icon-follow-comment icon-googleicon-hamburger icon-imedia-blog icon-imediaicon-instagramicon-left-arrow icon-linked-in icon-linked icon-linkedin icon-multi-page-view icon-person icon-print icon-right-arrow icon-save icon-searchicon-share-arrow icon-single-page-view icon-tag icon-twitter icon-unfollow icon-upload icon-valid icon-video-play icon-views icon-website icon-youtubelogo-imedia-white logo-imedia logo-mediaWhite review-star thumbs_down thumbs_up

How to clean your dirty data

By 2020, it's predicted that the customer experience will overtake price as the key brand differentiator. Astute marketers recognize that great brand experiences are rooted in good, clean data. After all, how can you target someone with a great offer if you don't understand their needs, behaviors, and how to reach them? Data needs to be both accurate and actionable to drive relevant, favorable, and valuable customer engagements that lead to stronger, long-term relationships.

Marketers of the past know all too well the weight of massive, disparate data sets, and the lag of 90-day refresh periods. But as we witness the modern consumer dart across channels at lightning speed, leaving a veritable breadcrumb trail of clues for us to follow, it's important to figure out what data you need, how often you need it, and how to keep it clean. Below are a few practical ways for marketers to effectively capture and update the valuable morsels of customer data while stripping away the sludge that weighs them down.

Start with email

Good email hygiene is a great place to start a new data regiment. With higher and higher hurdles to get your email delivered by ISPs that scrutinize email senders with increasingly sophisticated rules and tools, you need to focus here first or risk losing ground in this critical channel.

Familiarize yourself with ISP email basics
Did you know that ISPs collect millions of old, abandoned email addresses and use them as a means to score and monitor the quality of emails coming from IP addresses? Your IP address is like the social security number of email marketing -- score poorly and it's the same as having a bad credit score. Once you are identified as a potential source of spam, you could be blocked or inhibited in a number of ways. At the highest level understand that it's important to segment your customer lists by level of activity and validate the good email addresses. Take the less active groups and develop customized methods to test whether or not this email is still valid, or could be an ISP spam trap. Email addresses that do not pass your test should be suppressed and this group of customers should be tested in other types of outreach or campaigns.

  • Example of a spam trap test: Take customers who have been unresponsive to your email campaigns for 6-12 months and send them a highly aggressive offer -- "We've missed you! Come back and take 85 percent off any item!" Those who respond are reengaged with your brand and those who don't are either bait in a spam trap or not receiving your messages any way. Either way the email address in question is not the best way to reach them.

  • Develop clear bounce policies: Ensure hard bounce email addresses are immediately suppressed in your database and establish conservative criteria for soft-bounce emails, as well. If an email does not get through after three sends, your message is not being heard, consider this a hard bounce.

  • Know the dangers of rented lists: Some rented lists can be rock solid, others are downright dangerous. Your best bet is to develop your own permission-based email collection program and build a list of customers who are engaged with you and interested in receiving your communications. If you are considering renting a list, ensure that it is 100 percent opt-in and comes from a reputable vendor. Be aware that customers also change or consolidate email addresses more often than you might suspect. Ensure your email addresses are updated regularly and clean house on a monthly basis.

Update critical, actionable data sets -- often

In order to serve up the best possible brand experience, you need to update critical information sets in real time or near time. Information like what your customers are buying, what they are saying, and how they are currently engaging with your brand are some of the most important breadcrumbs you want to collect on a daily basis in order to drive truly informed brand interactions. Here are three examples of how these data sets, when collected in real time, could lead to stronger brand engagement.

  • Transactional data: If John buys a snowboard on your retail site and then walks into your store an hours later, could your sales person or an iBeacon message congratulate him on his new board, and offer him bindings, gloves, or boots at a discounted rate?

  • Social data: If Suzie posts how much she loves her new sweater to all of her 1,500 Facebook friends, and then goes to dinner near your store and tweets about her dinner plans, can you send her an email letting her know that for the next three hours she deserves $50 off her next purchase of $100 or more for being such a great customer?

  • Campaign response data: If Bob buys fishing line from you using a printed coupon from a direct mail campaign, and none of your 10 email coupons to him have ever been opened -- do you think you should continue to email Bob? If nothing else, perhaps it's a sign that a sales person should ask if he has updated his email, or prefers other methods of correspondence.

Free up the gridlock

Marketers need to get their hand on every bit of relevant data that they can use to enhance their customer's experience. But that should not be confused with getting their hand on every bit of data. Where the last 12 months of web behavior won't help guide your efforts, a summary profile of web behavior that could be turned into a persona would help significantly.

  • Clear the excess: Your approach to storing data should be to ensure that it is aggregated and stored at the summary level, where incisive questions are answered with precise metrics that can lead to clear, goal-oriented campaign directives.

  • Know your technology: Data warehouses are generally not structured with marketer needs in mind. For this reason, it is critical for marketers to capitalize on advances in database warehousing management technology and learn how to extract only the data they need.

Unleash the power of unstructured data

It is widely accepted that roughly 80 percent of business-related data is unstructured, meaning it exists in a form that is not a structured table with rows and columns -- but may contain text (e.g., emails, reports, social media posts), images (e.g., pictures themselves or slide shows), video files, etc. Before proving useful to marketers, unstructured data needs to be turned into insights. Here are two examples of how unstructured data can be put to use:

  • Use social to test real-time: Social media is an example of a massive source of unstructured data. If you are a sneaker company, you can scan the twittersphere for tweets looking for sneaker recommendations. Take those twitter handles, cross-reference that post with your customer database, match it to a Twitter profile, and use the email address associated with the Twitter profiles to send a targeted offer -- including reviews -- that satisfies your customer's need in real or near time.

  • See patterns and identify opportunities: Analyzing unstructured data can help reveal patterns of behavior over time. Do customers who provided negative survey information on your cable TV services check in on a social platform when they go to the movies? If you are a cable company, can you use that information to offer some great, new movies on Friday nights to anticipate your customer's moments of need? Often times marketers don't need all of the underlying data -- what they really need is the ability to see a pattern in time to take action.

Ultimately, the dirtier your data, the less accurate your marketing efforts become. And when you deliver a poorly targeted message in a channel a customer no longer cares about, your efforts -- and ability to capitalize on them -- is in vain. Marketers need to examine the health of their database beyond deliverability and opt in or out. They need to understand what types of information their customers feel comfortable sharing and deliver highly customized experiences across multiple channels -- showing respect for who the customer is and how they want to be communicated with. This capability to market with accuracy and agility is going to be a distinguishing factor for the marketers of the future and the key to tapping into the needs of your customers at decisive purchasing moments.

Paul Mandeville is the chief product officer for QuickPivot.

On Twitter? Follow iMedia Connection at @iMediaTweet.

Paul Mandeville is the Chief Product Officer for QuickPivot, the leader in real-time cross-channel marketing automation technology and services. Paul has been an innovator and entrepreneur in the marketing automation technology industry for over 15...

View full biography


to leave comments.