Data management... not at all sexy, but, hugely important
Let's face it data management is a term that doesn't get the blood flowing quite as much as trendier, sexier jargon like marketing automation, personalisation and programmatic advertising. But the precise management of data is at the heart of all these marketing techniques.Modern databases tend to be dynamic. The data in them are always evolving, being edited, added, deleted, tagged etc. Because of this there is the potential for bugs to creep in from time to time, especially if data can be manipulated by employees and end users. So why is it important to keep your database integrity intact? The short answer is that is can end up costing you more money when it comes to marketing activities. Here are some examples of how the knock on effect of poor data can generate unnecessary cost.
1. Duplicate email addresses on Mail Chimp
A lot of Email Service providers use a pricing model based on the volume of emails you send per month. With Mail Chimp, after you surpass your free 2000 subscriber limit, you start paying by the volume of email addresses stored within your account. Not by unique email address, by email address. So if you have multiple lists which all feed into a larger 'master' list, which you use for emails then you'll have duplicates in your account. In most cases it's far better to use a single list and segment customers using list fields or groups. Also note that you can't delete an email address until at least 7 days after any email has been sent to that email address. So if you're sending emails from multiple lists and want to consolidate your lists into one, you may have to wait a week to do this. Plus if you're using multiple lists to send campaigns from and you ever change provider, you have more work to do to keep on top of who unsubscribed or bounced from each list. One list = no duplicates, and far easier to manage short and longer-term.
2. Bogus email addresses
Be mindful of adding email addresses to your master email list if they've been collected from a separate source, especially a form linked to a competition or free sample. Some people like to enter weird and wonderful email addresses when they sign up for these types of things, especially those trying to game the system. Adding bogus email addresses to your master list will add volume to your database, but not quality. As referenced in point one, if you're using Mail Chimp, volume costs money, so you don't want any email address on your account which isn't a 'proper' address.
Bogus email addresses can also effect account and domain credibility. If your list contains a lot of bogus addresses and you attempt to send them emails you might get a high bounce rate. This, in the worst case scenario can cause your Email Service Provider to close your account temporarily if you send your emails through their domain (standard for most ESPs). Not handy if you've got a time-sensitive email to dispatch. If you route emails through your own domain using a custom DKIM then there's potential for your domain to tarnish its credibility too and you may struggle to reach certain inboxes in the future.
3. "No longer at this address - return to sender"
This is a classic case of out-of-date data. People move house. They'll tell their bank and utilities companies but rarely will they tell an online retailer until they come to order again and change their billing address. For those direct commerce companies who send catalogues out to targeted groups of customers it's important to ensure addresses are up to date otherwise you're just adding unnecessary print, postage and fulfilment costs.
4. Understating the value of customers when they're treated as duplicates
If you are performing analysis to determine which customers are more valuable than others, you might omit customers from a segment if they have been duplicated. The knock on effect depends on the actions taken with either the segment of valuable customers or the remainder of the database. If one segment is given a thank you reward offer and customers miss out on it who should be eligible, then that's a missed customer engagement opportunity. Alternatively the remaining, less valuable customers might be targeted with a deeper discount offer to get them to buy again. This has the potential to subsidise purchases which might have been made anyway or with smaller discounts.
Reporting accuracy gets thrown out when duplication occurs. Customer Volumes, Average Spend Per Customer, Customer Lifetime Value and Repeat Purchase Rate can all be impacted. In the worst cases this can lead to Marketing teams wasting time and money trying to work out why certain KPIs have increased or dropped, when they actually haven't.
5. Confusing communications
Ever received a communication from a retailer that baffled you? Most of the time the confusing communications we receive boil down to a lack of relevance as a result of loose or no targeting; not so much a lack of data management but a lack of data mining. But sometimes poor data management does come into play. Take the scenario when you still receive mail through the post, despite ticking the box that says don't send it. More unneccessary cost for the sender. This is a simple case of data either not being updated correctly or systems not talking to each other coherently. Setting preferences within a user account seems to be where the issues occur most based on our observations. We can reference several multi-million dollar online and multi-channel retail businesses that don't change what they do despite customer preferences being inputted. The cost here being one of customer frustration and loss in credibility.
Data Maintenance and Data Cleansing
So in order to keep a quality database, it is good practice to put in place processes which maintain its integrity and which cleanse the data. Data maintenance is the ongoing process of keeping data fit for purpose. It can involve processes which remove duplicate entries for customers who have shopped as a guest or with their account, so that their activity is recorded under a single user. Thus, a Single Customer View is a great example of a data asset which maintains the integrity of a retailer's customer database.
Data cleansing tends to happen periodically and there are some great, affordable tools to use which can help remove bogus emails which integrate directly with common ESPs like Mail Chimp. Data Validation is one, we particularly like and have used with our clients. At the time of writing this article it cost about £35 to cleanse a list of 10,000 email addresses. For those retailers wanting to do a mailshot address cleansing is worthwhile for anyone on your database who hasn't been active for at least six months. It is worth passing their address data through the royal mail's National Change Of Address file, using a registered data bureau like The Data Processing Company. The Data Processing Company can also help with suppressions that can remove deceased customers and those on the mailing preference service who have opted out of receiving marketing through the post.
Some summary Tips on data management:
Manage email lists pro-actively. Mail Chimp accounts with multiple lists need a keen eye on them and a simple process to ensure duplication is minimised.
Get a Single Customer View that continuously maintains the integrity of your customer data allowing you to act upon it with confidence.
Cleanse email address data at least once a year, and postal address data at least once a year, especially before any catalogue mailings.