Melissa Data Quality Review

On-premise Data Quality components ensures accurate mailing campaigns

What is most valuable?

We have a large volume of customers (millions) and this suite is the best priced for what we need.

We use the cloud version for NCOA and the on-premise version for the normal batch processing from our POS systems. The two components we use are the Contact Verify and MatchUp components.

Contact Verify: Ability to cleanse the name, phone, & address information. The codes that accompany the output record helps our system determine whose address is accurate and if not, what are the issues with the address. This software assists in more complete analysis of our customers.

MatchUp: Ability to "DeDupe" and retrieve possible customer matches which keeps our customer master clean and accurate.

How has it helped my organization?

Before we used DQSand an algorithm for matching and only updated valid address information quarterly. This caused returned mailing and upset customers. This suite has monthly updates to ensure we are up to date with recent new home construction and gives us the ability to modify the matching rules to our needs. We have seen a reduction of calls and returned mail.

I was referring to two different components: Contact Verify and MatchUp.
Contact Verify: Our company runs campaigns for our membership and needs accurate address information. We previously "cleansed" and matched the incoming customer addresses but did not necessarily know if the address is valid until we ran our quarterly National Change Of Address process. Melissa Data's contact verify component is updated monthly with national addresses. When the address runs through the component the output is not only "cleansed" but tells if the address is valid (Cass certified) or if there is some other issue with the address.

MatchUp: The matching component has rules that can be modified to look at address or combo of columns to match.

What needs improvement?

Not in order of importance:

1. Ability to multi-thread. Currently we have to split the data into different paths to try to quicken processing time. In all fairness, we do have a very large customer master to match against.

2. Data types default to ANSI. If you add a column and you have everything else set to Unicode, you sometimes need to either redo the component or do a type conversion step after the output.

2. The ability to resize the selection windows in the components to see the entire field name.

3. Although you can get the component running rather quickly, to do the more complex matching takes more knowledge of the tool. Documentation & videos on the on-premise version is not very helpful. It takes some trial/error to see what the tool is actually doing.

4. License keys gave us problems this past year (previous years were fine). We were sent expired keys and keys that did not match our components. The environment variable does not always pick up so to ensure non-interruption the key needs to be typed in the configuration text files as well.

5. We log SSIS errors and information to assist in abends and audit. We capture only the 110 & 120 messages. If every output stream pin in the MatchUp output filter is not assigned, the SSIS log file fills with "110: Household Process:Warning: The output column "cass_first_name" (49) on output "OLE DB Source Output" (43) and component "Existing Households" (32) is not subsequently used in the Data Flow task. Removing this unused output column can increase Data Flow task performance." for each unused output stream.

For how long have I used the solution?

Three years

What was my experience with deployment of the solution?

Normal SSIS deployment of packages that contain components. The Melissa Data software itself, however, takes a lot of space on the hard drive so be aware of the space needs. There isn't any real clear instructions on doing an upgrade (it does not seem to delete the older, no longer needed files). Also make sure you have enough memory.

What do I think about the stability of the solution?

No issues with stability

What do I think about the scalability of the solution?

We had found a bug that only appeared when trying to match over 94,000 records. The process hung and we could not identify without a lot of testing what was going on. Their support team worked with me to determine the issue.

How are customer service and technical support?

Customer Service:

We get timely update notices and address data file notices. Very responsive if an issue or question arises.

Technical Support:

Technical support during normal work hours has been very good. The developers are very attentive and will answer questions quickly.

Whenever we find a bug their support team diligently works with us to determine the issue and to get it fixed. It is for this reason I agreed to assist them in rating their software.

Which solution did I use previously and why did I switch?

DQS / Custom Algorithm - limitation on number of customers it can process and stale data.

How was the initial setup?

Documentation & videos on the on-premise version is not very helpful. It takes some trial/error to see what the tool is actually doing.

What about the implementation team?

We have in-house SSIS experts.

What was our ROI?

More accurate address information, less returned mail, less customer calls.

What's my experience with pricing, setup cost, and licensing?

The pricing is very competitive and the product is robust enough to get the job done. Be sure you know what you are licensing. There are components that have similar functions but may not be what you need. As mentioned, we have everything we need in just the two components.

What other advice do I have?

We were not thrilled with the "phone home" requirement as we are skittish about opening a port and allowing 3rd party software to pass files. Luckily we found out it was required for the Geocoder component which we do not use. See link below for more information:

Disclosure: I am a real user, and this review is based on my own experience and opinions.
author avatarGaryM

Agreed on all points - especially the need to multi-thread - I also had to split up our processing due to this to speed it up. However even so it still appears much faster than any other similar tool I've used in the past. And YES regarding resize! That's SO annoying. No other SSIS component has such a restriction. And yes the license thing bit us also. Love having production process stop working and in a panic have to call the vendor. Really, was that necessary? But then if they were a perfect vendor and product we'd have nothing to complain about ;-)

author avatarGaryM

An update for out when you upgrade SSIS to 2016, it's not fun. Also I split matching into multiple dataflows and this slashed runtime by about 75% (from 4 to 1 hour for full exact match with advanced survivorship).

Sign Up with Email