What are potential dupes?
In any box office system it’s inevitable that there will be duplicate records, customers who for one reason or another have more than one account on the system. It’s also possible that there will be customer accounts that look as though they are duplicates but actually are not - customers with the same name or initials who live in the same postcode, for instance.
Potential dupes then are accounts which have the potential to be duplicates - i.e. they might be unique accounts, or they might be duplicates.
How does Spektrix deal with these?
Spektrix looks at the surname and postcode of every customer record and applies a flag to any account where these pieces of information match another account. This is called the Potential Duplicate metric, and is the system’s way of identifying that a Customer Record might be a duplicate - the criteria being customers who have either the same initial, surname and postcode, or the same initial, surname and first line of address. Of the accounts in your system which have this metric, some will be genuine duplicates while others will, upon closer inspection, prove to be unique accounts.
How can I deal with potential dupes?
With the use of the Potential Duplicate metric in Spektrix you can easily identify all accounts which the system considers to be potential dupes, and then check through each account to decide whether it is a duplicate or not. It’s quite a manual process but if you can get on top of it then you’ll find it’s quite simple to keep your database nice and clean.
The first thing you need to do is to create a Customer List to pull out all of the accounts which have the Potential Duplicate metric. You can do this in the Marketing Interface under the Customer Lists section.
- Create a new Customer List and name it something along the lines of Potential Duplicates.
- Create a new Local Segment - it needs to be a Customer segment.
- Expand the Individuals wrapper and drag Potential Duplicate down into the criteria section.
- Make sure you tick the box next to Potential Duplicate - this defaults to not being ticked.
- Save your segment, then drag it into the criteria section.
- Make sure you update the count on your Customer List so that you can see how many records it brings up. This is how many potential dupes you have.
- Add in Customer ID, Name and Billing Address Postcode into the Output of your Customer List.
- From the Choose an action drop-down list, pick Output as CSV and save the file to your computer.
Once you have your Customer List with all of the potential duplicate accounts you need to start looking through the accounts to identify whether each one is actually a duplicate or not.
Working off the Customer List and checking on your system, you should be able to quite easily spot which accounts are duplicates and which are not. Any duplicate accounts then need to be merged together so that you end up with just a single account per customer.
If you’re not sure how to merge customers, have a look at this article. You can only merge two accounts together in one go, so if you find customers who have three or more duplicate accounts you’ll just need to merge them together in stages - pick two to merge together first, then merge the third account with the newly-merged account, and so on.
Facilitated Booking
If you use the Facilitated Booking feature, you will need to create a slightly different Customer List, as the Potential Duplicates metric won’t pick up any accounts created from the Choose Attendee page online. That’s because these accounts are created without any address data, and the metric is based on looking at postcode data.
In addition to what you’ve already set up, you will need to add a Customer segment looking for Individuals without a billing address, and whose accounts have been created over whatever period you’re working to. In the example below we’ve set this to Last Week, on the assumption that this will be run weekly, but it will depend on what point during the process you’re doing this.
- Make sure to leave the Has Billing Address option unticked, so that it looks for Individuals who do not have a billing address.
Once that’s done, add both of the segments you’ve created into your Customer List.
You can then continue with the overall process as detailed in the section above.
Next steps
Once you have gone through the full Customer List, merging duplicate accounts together where necessary, you should have cleared the vast majority of the dupes on your system.
We recommend that you continue to regularly check for potential dupes after this point. Your list of potential dupes should be much smaller any time you run it subsequently, so it will be much easier to keep on top of any newly-created accounts that might be duplicates.
If you set this as a regular task, you can schedule the Customer List to automatically run at regular intervals. If you’re not sure how to schedule Customer Lists, see this article for more information.
You can then amend your Potential Dupes segment in the Customer List to look for new accounts created over a certain period.
- Edit the Potential Dupes segment.
- Expand the Individuals wrapper and drag the Date Created metric down to the criteria section.
- Choose a date metric that corresponds to how often you plan to run the Customer List - i.e. if you are going to run it weekly choose the Last week metric.
- Save the segment, editing the name to reflect the date metric.
- Create the schedule for your Customer List, setting it to recur every week/month etc.
That will then allow you to incorporate de-duping into your regular tasks, making sure that you keep your database nice and clean.
In summary
Keeping on top of potential dupes is undoubtedly a manual process, especially at the beginning if there is a large backlog. It’s an important task however, so getting into good habits as early as possible is worthwhile.
If you have any further questions about this process, or would like to discuss it in more detail, please don’t hesitate to get in touch with the Support department who will be able to help you with anything else you need.