This is one article of a series appearing in this issue, categorizing the users of the corporate information factory.

What comes to mind when you think of farmers? Old McDonald? Straw hats and tractors? Hardy, diligent folks who put food on our tables? You bet. They cultivate their soil, sow their seeds, pray for good weather, harvest their crops, go to market and plan for repeating the cycle the following season. Although a simplification, it's remarkable how well the farmer description fits certain corporate information factory (CIF) users.

See if this sounds familiar. CIF farmers have:

  • Well-defined requirements: their crops.
  • Consistent requirements: limited crop rotation.
  • Data to create the answers: their fields.
  • Clear, concise queries: their seeds.
  • Access tool expertise: they know how to drive the combine.
  • Sound business understanding: they know what their crops are worth.

Farmers usually come from the management or business planning groups within your organization. Typical farmers in your corporation may be the financial analysts responsible for reporting on revenues and costs. They may be sales and product analysts determining how well a product is selling in some part of the world. They may be the people who track campaigns or promotions from week to week or they may be the analysts who monitor the budget versus actual reports.
Wherever they are in your organization, you can bet that they have a very good handle on the types of reports and analyses that they need to perform. They are a joy to data warehouse implementers.

Understanding Farmers' Needs

Well-defined requirements: corporate information factory (CIF) farmers have well-defined requirements; they know what they want and can generally express it clearly and concisely. For example, people who analyze product profitability or sales reports each month know exactly what data they want, how they want it displayed, when they want it, what media they want it in, etc.

Consistent requirements: Their requirements rarely change or change very slowly, giving us a solid basis to begin growing our strategic decision support environment. The unwavering nature of their requirements gives us the chance to stabilize the environment, understand the iterative process, set up appropriate technologies, establish proper processes and develop infrastructure for the more difficult and less stable requirements of the explorers, miners or tourists.

Because their queries are predictable and routine, they also expect good response times. It is not unusual or unreasonable for the farmer to expect sub-second to a few seconds response time on a report run every Monday morning at 8:05 a.m. Since we have been forewarned, we can set up the database to facilitate these queries.

Data to create the answers: Farmers have been struggling to capture solid, strategic information for a long time without the CIF, so they understand the value of integrated, cleansed data. They tend to be some of the most satisfied users of the CIF because their requirements are so easily met and managed.

They look to the CIF to supply automation (of reports), consistency (of data), reliability (of the technology), standardization (of calculations and algorithms) and reproducibility (of information).

Clear, concise queries: Farmers in the DSS world ordinarily submit repeating queries. What are the total sales for product A in market segment B over C time frame? How many customers bought product A through sales channel B during campaign C in the last month? They can also tell you the parameters they will need to run these types of questions. And they submit these queries with precise regularity (e.g., every Monday morning, every month end, every end of quarter, etc.).

Another characteristic of farmers' queries is that the query is generally very short and retrieves a small number of rows from the database. Rarely are they asking for thousands of rows of data or years of history. A typical query may involve only the recent month's data. Many times the farmer is interested in only relatively current data ­ sales from this month, for example. They may do simple comparisons as well ­ this month to last month, this quarter to the same quarter last year, etc.

A final characteristic of farmers is that they almost always get an answer to their query. Unlike the miner or explorer who may or may not get an answer to their question, the farmer knows that he will get something back. It may not be what he expected or wanted to see, but it will be a response.

Access tool expertise: Farmers tend to be relatively knowledgeable when it comes to the manipulation of data for reporting purposes. For years, they have been on their own, building reporting databases to garner the information they wanted. Therefore, they readily adopt a new data access tool and easily begin using it.

Sound business understanding: These end users create basic business intelligence reports that they run routinely, glean information from, study trends for the future, and then report their findings back to the corporation. They are also, without doubt, the largest population of users of the CIF.

The farmers in the business community track key performance metrics that report on the health and well-being of the corporation. These types of standard reports include those that show the ROI of major decisions (e.g., opening a new store, running a new campaign, offering a specific mix of products), the costs and revenues associated with doing business, balanced scorecard reports, etc.

Many times, the farmer provides feedback on the effectiveness of an explorer's predictions. The explorer makes predictions based on patterns discovered in the exploration warehouse. For example, the explorer may predict that sale of a particular product may show a 50 percent increase in a certain segment of the country. It then falls on the farmer to track that prediction and to determine whether it is indeed increasing as predicted.

Architecture Implications

The farmer sees the world in terms of both dimensions (such as product, market segment, campaign and sales channel) and metrics (such as revenue, cost, counts, transactions, etc.) Therefore, the farmer's DSS environment consists of data marts found within the corporate information factory (see Figure 1). Each data mart is created from the known requirements for a particular function or capability1 (product profitability, sales channel analysis, etc.). Typically, these marts are multidimensional in nature using either relational or multidimensional technologies. The slice-and-dice world of these technologies fits the usage patterns of farmers quite nicely.

Their database design is a star or a snowflake schema if relational technology is used. If a multidimensional OLAP tool is chosen, a hierarchy of the dimensions along with the aggregated facts is determined for the predefined and pre-made cubes. Again the ease of access and response times associated with these technologies fit the farmer's world quite nicely. Should a new report, cube or query be needed, they often turn to their line of business or central IT person for help.

Their data is usually aggregated or summarized to a fairly high degree. Since the farmers know what they are looking for, they rarely need to see massive amounts of data to find it. Nicely aggregated and summarized data gives them all the information they need to understand whether a product is doing well or as expected, whether a sales region is performing according to plan, whether a channel should be eliminated, etc. They may need to drill down one or two layers of detail within the data mart they are using, but rarely will they need to drill down to the lowest level of detail found in the warehouse. Because of this characteristic, the farmer's access to the data warehouse itself may be restricted.


Farmers represent a substantial percentage of CIF users. They are often proponents of initial CIF projects, and their analysis is extremely important to the ongoing health and well-being of an organization. Farmers have clear, consistent requirements that are frequently satisfied by standard, repetitively executed queries. As a result, CIF repositories can be implemented and tuned to optimize their queries and deliver improved response times.

By implementing the CIF to support farmers as described above, we can efficiently and effectively meet their needs by delivering the data:

  • that answers their queries,
  • in the right amount,
  • at the right level of detail, and
  • in a timely fashion.

1 See "Will the Real Data Mart Please Stand Up?" by Claudia Imhoff. DM Review, March 1999.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access