Data In - Or Data Out

Database management is a complex process that involves highly-trained professionals from multiple disciplines. Some companies perform database management in-house, and others have made the decision to farm it out. Which is right for you? We will discuss a number of factors that can influence your decision.

Different Levels of Outsourcing

The outsourcing model comes in two basic permutations, with gradations in between. In the first, everything is outsourced. This includes IT infrastructure (hardware and software), associated technical people, and data leveraging personnel (CRM experts and data miners). Optionally, the client is provided with access to the database; access that generally is limited to viewing cached reports and running canned queries.

In the second permutation, only the IT infrastructure is outsourced, including the robust connectivity required for complete client access. Here, the client drives all of the strategy and tactics, including construction of statistics-based predictive models, execution of analyses, and generation of selects for mail and email.

Choice of Partner: Critical for Outsourced Success

With an outsourced solution, most of the risk lies with the choice of partner. This is especially true with data quality assessment and assurance, and with the leveraging of the database via CRM experts and quantitative analysis.

Do Not Be a Small Fish in a Big Pond

Often, trouble results when a database management company wins an account that is smaller than it usually takes on. This is common during the scramble to meet revenue objectives. After the deal is closed, it is tempting for a vendor to assign the "B" team to do the work. So, be wary if you represent "say "$300 thousand a year in revenue to a vendor with mostly upper-six-figure clients.

The Parable Group found this out the hard way in 2000. In the words of Tim Blair, Parable's Executive Vice President & General Manager: "Our database was maintained by one of the best-known information and CRM companies, and they also did our mail processing. We naively assumed that we were in good hands at this large and visible company. Once we started digging into our data, we realized just how wrong we had been. We were overwhelmed at its poor condition."

Especially Avoid Being a Small Fish with Complex Data

Parable was also faced with this problem. It produces catalogs, flyers, web sites and promotional kits for 242 locally owned and independent Christian bookstores throughout the United States and Canada. Therefore, the database must reflect hundreds of different stores and store chains. The sophistication of data capture varies dramatically across each, and three different point-of-sale systems are employed.

Another complexity is that each standalone store and store chain must be treated as a separate profit center. If someone has purchased from multiple standalone stores or chains, then the database and corresponding reports must reflect each relationship. For example, a customer might be forty miles from Store #1 in Chain A, but only three miles from Store #2 in Chain B. Or, he might have purchased thirty months ago from Store #1 but two months ago from Store #2. The only interest the owner of Store #1 has in the customer's close proximity to, and very recent purchase from, Store #2 is the extent to which they are predictive of future purchases from Store #1.

Likewise, business-to-business direct marketers face notoriously complex data. The standardization, consolidation and leveraging of this data requires deep and specialized expertise. However, business-to-business databases tend to be limited in size, with relatively modest circulation quantities. Generally, this makes them unattractive to the larger database management companies.

Is Database Management a Core Competency?

Parable realized that offering cutting-edge database management to its client bookstores would provide it with a compelling competitive advantage. Therefore, the decision was made during the later half of 2000 to bring everything in-house. The process was long and arduous but ultimately successful.

If you decide to outsource, make sure that database management is a core competency of your potential partner and not an "add-on" service offering. Unfortunately, there is some truly awful work being done in our industry. This is because advanced hardware and software are readily accessible. However, the danger of misuse is great when individuals with sub-par analytical training are armed with powerful tools to build, maintain, manipulate, query and mine databases. False insight is worse than no insight at all! For example:

In 2003, a direct agency constructed a database for one of its clients and then employed a leading OLAP tool to perform extensive analysis. Unfortunately, the resulting "customer snapshot" report was rife with error. One highlighted "crisis" was a dramatic decline in the percentage of customers who were ordering a second time. However, this was an artifact of failing to control for time-on-file. The rate of second-purchase is always lower for customers who are relatively new-to-file because they have had less time to reorder!

Tight Integration with Operational Systems

Some databases are standalone entities. They receive transaction feeds from operational systems, and then optionally return enhanced and standardized data. These circumstances facilitate an outsourced approach.

Other databases are tightly integrated with operational and CRM systems. They may, for example, feed customized screens to assist Telemarketing Sales Representatives. Or, they may supply key information to retail point-of-sale systems for use by associates while ringing up purchases. The corresponding logistics make an in-house approach attractive to many companies.

Concern for Privacy

Certain types of data such as medical history are intrinsically more susceptible to privacy concerns. Often, companies with sensitive data opt for an in-house solution. An example is a well-known direct marketer of products geared to kids. It decided that a security breech by an outside data management company would be a public relations disaster.

Personnel Redundancy

Employee turnover is a fact of life. Therefore, an in-house solution is risky unless you are large enough to support overlapping personnel for critical job functions. Data mining and mail processing are particularly vulnerable skill sets. Many smaller firms have been hurt badly when a key database management employee moves on to greener pastures. Generally when this occurs, much "company memory" is lost forever.

Another problem with a small number of database management professionals is that people thrive in a collegial atmosphere with clear paths for advancement. When these are absent, employees are more likely to start interviewing with other firms.

Company Politics

Generally, IT favors an in-house solution and wants to assume ownership. There are three problems with this:

First, an IT organization inevitably has at least 100 tasks on its to-do list at any given time. With perpetual resource contention, the needs of the marketing department will invariably be deemed a lower priority than mission critical activities such as inventory management and payroll distribution.

Second, IT often becomes enamored with bleeding-edge technology. The result can be an over-engineered solution that takes forever to implement. For example, one prominent company currently is going live with its database after several years of construction. Over-engineered solutions also come with substantial price tags that make it difficult to generate positive ROI.

Finally, many of the skills required for database management fall outside of the realm of a typical IT department. Consider business-to-business data, which was discussed earlier. It is a profound challenge to create, maintain and manage the complex linkages across individual contacts, sites and overall company entities. All of this requires a deep knowledge of business-to-business data, direct marketing practices, and hygiene and consolidation technologies. The same is true of the data mining required to support ongoing target marketing.

A Hybrid Solution

A "rent-to-own" hybrid solution might be right for you, where you have the option to move the work in-house at some future date. With this approach, you agree to outsource your database management for a minimum of "say "three years. Typically, there is a clause in the contract to handle the transfer of database management in-house, should you eventually decide to do so. That way, you can get the project up and running quickly and then decide later what you want to do in the long term.


Is an in-house or outsourced approach right for you? The answer lies with the specific circumstances that surround your company. If you go with an in-house solution, make sure not to skimp on the investment required to build a deep and seasoned staff. If you decide to outsource, then be mindful that your ultimate success will depend on the selection of an appropriate partner. If you are unsure which way to go, think very seriously about a rent-to-own hybrid solution.