Benchmarking is an inevitable part of any production process. It can be done informally, through comparisons during conversations at meetings or other get-togethers. It can be done through retrospective comparisons, by asking producers to recall prior productivity levels. Finally, it can be done by taking the records and analyzing the data in a standardized method across farms.

The latter is the method used in this benchmarking report. The broad use of PigCHAMP recording on farms along with the wise decision, made many years ago, to benchmark based on the collection of actual production data has created an incomparable opportunity to compare and measure productivity on swine farms.

IDENTIFY TRENDS
In this database producers submit, on a quarterly basis, a backup of their production data, which is then used to create a quarterly report back to these producers to allow regular benchmarking. Such benchmarking can identify trends that deviate from expected productivity and can produce real opportunities for improvement. For reproductive records, quarterly comparisons are approximately the maximum frequency that is of use.

On a yearly basis, however, this database affords the opportunity to identify longer-term trends. In addition, the yearly review allows partners in production to recognize opportunities for improvement in the current herd. As always, there are illustrations that present opportunities, as well as the limitations of current productivity outcomes. Some are universal to the industry, and some are limited to underperforming farms. As you review these results, it is useful to keep a number of aspects in mind.

• This database summarizes performance across individuals sows or growing pig groups, and then summarizes on a farm-by-farm basis. This allows data validity steps to be taken to ensure utility. We test whether the sows' productivity is recorded correctly and whether herds are stable and in production methods. This level of validation is often not seen in summary results in agriculture.
• Most of these comparisons are in the area of reproductive effi ciencies. Reproduction benchmarking is often characterized by differences in capability to achieve high performance. That high performance is rarely tied to any single costly technology. In most cases, the variation that we see in reproductive productivity is not a strategic differentiation but simply an outcome of management. In other words, most farms are capable of reaching the higher levels of reproductive effi ciencies seen in this database.
• We are seeing generalized trends in improvements in reproductive output, and we believe the long history of PigCHAMP benchmarking has contributed to these improvements. There are two ways of viewing causes of improvement. The first is to suggest that improvements in productivity have occurred in all farms. The second is to suggest that there is a "selection of the fittest" occurring, where only the best farms survive. Though the latter does occur, it appears that most improvement is occurring within farms.
• It is still surprising that there is such a large range of productivity. There are very few technological secrets or capabilities that are not available to all participants in the industry. Yet we must view opportunities for improvement as relatively difficult to attain. It is unlikely that there is one "silver bullet" that can be discerned from these benchmarks for a farm.

Editor's Note: Dr. John Deen is a veterinarian and associate professor at the College of Veterinary Medicine, University of Minnesota.

To improve herd productivity, producers need standards and targets for key production parameters to compare with their own performance. Comparisons should be made using same calculations and formulas.

Benefits for Participating Producers:

• Obtain our standards and targets to compare their performance using the same software and measurements
• Research that delivers useful information to swine producers, and practicing veterinarians provided by the University of Minnesota
• Evaluate their farm against industry standards and help them in setting performance goals
• Provides percentile rankings for each farm value based on comparisons to other eligible herds for continuous improvement
• Provides summary reports free to PigCHAMP licensees

Yearly benchmarks serve as "state-of-the-industry" reports that provide both motivation for change and recognition for how far the industry's come. The PigCHAMP annual year-end benchmark, for example, reports average production values as well as the upper and lower 10th percentile values for participating sow farms.

But what does it all mean? For pork producers, the annual PigCHAMP benchmark data and analysis offer points for herd comparisons. In many ways, the annual benchmarks are like new technologyÑtheir value lies less in the possession and more in the implementation. And that leads into benchmarkingÑthe active use of benchmarks to improve productivity.

Looking at relationships within the sow productivity data provides an opportunity for "ecological" benchmarkingÑlarge-scale exploration of factors associated with more and less productive operations. However, ecologic benchmarking has several challengesÑbeginning with ecologic fallacy. The ecologic fallacy is that relationships observed at the level of the group are not necessarily true at the level of the individual [farm]. In other words, it is not hard to find an exception to the "rule". Further complicating the issue with farm-level summaries is that average production values represent averages across all farms and do not take into account differences such as herd size, health or genetic stock.

It is possible that the annual PigCHAMP benchmarks are a biased representation of the industry since herds that use paper or other record-keeping systems are not considered at this time. This could also mean that performance of herds with poorly kept records or poor performance is excluded from consideration. Farm-level bonus systems may lead to manipulation of certain performance parameters in generated reports and cause misrepresentation of actual performance. Incomplete event recording can cause disparities between actual and paper performance, and differences in event recording practices between farms can limit the ability to compare some measures.

Despite their limitations, the annual industry benchmarks continue to shape both producer and customer expectations. Seedstock suppliers are expected to offer both prolific and efficient animal lines. Nursery managers, recognizing the risks associated with lightweight piglets, have increased weight standards for acceptable animals. This, in turn, has led to longer optimal lactation length at the sow unit.

Although the content of quarterly and year-end benchmarks may be limited at this point, they provide relevant information. Current benchmarks serve primarily to set expectations. As with other technologies, the value of the benchmarks lies within the implementation of benchmarking and the steps taken toward herd improvement as a result of the findings.