Hospitality Industry Case Study

In most performance management solutions, key performance indicators (KPIs) are developed and implemented with targets and thresholds that appear to remain constant throughout the year. However, in the real world, KPIs will vacillate from month to month or even week to week due to the dynamics of business trends, cycles and seasonality. During the course of a year, monthly KPI levels can experience up to a 30% shift between peak and trough values due to seasonal impacts alone. In the hospitality industry, weather profiles, major holidays and vacation cycles can cause tourist demand to fluctuate dramatically from season to season and month to month. The bottom line is that KPI targets need to be adjusted to reflect this uneven customer demand across the year. This prevents the inevitable "roller coaster ride of emotions" as targets are exceeded or missed due merely to seasonal aberrations. This month we will investigate the simple technique of deseasonalization that allows us to translate "static" KPIs into more seasonal "dynamic" KPIs. The impact can be quite dramatic. In the following case study, KPI actuals from three of the months would have been misread as being below the KPI target level when, in fact, the actual KPIs exceeded the KPI target levels.

Development of Seasonality Indices

As I mentioned in the August issue of DM Review, standard metrics such as RevPar (revenue per available room), ADR (average daily rate) and occupancy (rooms sold divided by rooms available) have been developed to capture financial and operational excellence in the hospitality industry. Although we will specifically focus on the seasonal attributes of the occupancy KPI, the following approach can certainly be applied to both the RevPar and ADR KPIs, as well as consumer demand patterns in many other businesses such as food products, beverages, pharmaceutical and oil industries.

In our case study, we will illustrate how a fictitious hotel property, Hotel Dreamage in Hawaii, can incorporate seasonality into the calculation of their KPI targets to provide more realistic values that capture true month-to-month variation. The basic approach is to leverage three years of historical data to develop monthly indices. When the resulting index is greater than 1, the occupancy level is said to be above the KPI average target of 73.3%; when the seasonal index is less than 1, the occupancy level is said to be below the average target. The actual monthly occupancy level is then divided by the monthly seasonality index to create the deseasonalized monthly KPI occupancy metric. This value (rather than the KPI actuals) should then be used as the "true" monthly KPI target.  

The seasonality indices development and analysis process consists of the following steps:

Step 1:  Select KPI variable to be deseasonalized

  • Occupancy levels for Hotel Dreamage

Step 2: Determine monthly KPI target value

  • Monthly KPI occupancy target set at 73.3%

Step 3: Develop monthly indices

  • Sum all the monthly occupancy levels for the last 36 months (three years)
  • Sum the three values for each month to provide monthly totals
  • Divide the monthly totals by the 36 month total to calculate monthly seasonality indices

Step 4: Determine monthly deseasonalized KPI targets

  • Divide each month's KPI occupancy actuals by the monthly seasonality index

Step 5: Create prescriptive statistics

  • Calculate the percent miss vs. KPI target of 73.3%
  • Determine incorrect KPI value calls - KPI target "exceeds" that were called KPI "misses" and KPI "misses" that were called KPI "exceeds"

Seasonality Profiling and Analysis

The results from these calculations are found in Figure 1 and are segmented into four distinct sections which focus on the seasonality indices, KPI metrics (non-deseasonalized and deseasonalized) and the directional status. The first section, occupancy profile, reveals some rather interesting statistics. The three-year averages developed for the monthly occupancy levels show a peak of 84.1% for February and a low of 64.1% for September. Although, they will eventually average to the annual KPI target over the course of the year, each month is distinctly different from the annual KPI target. Once these occupancy values are converted into seasonality indices, it is readily apparent that seasonality is rampant throughout the year. The months of February (index = 1.173) and August (index = 1.110) exceed the baseline of 1.00 by 17.3% and 11.0%, respectively. On the downside, January (index = .897) and September (index = .894) fall short by more than 10%.


Figure 1: Hotel Occupancy Seasonality Profile

The next section (see KPI - non-deseasonalized) shows the KPI actuals without any adjustment for seasonality. In this scenario, the January KPI occupancy level is 18.4% below the KPI annual monthly target of 73.3%, while February's KPI metric exceeds the target by 22.5%. With seasonality incorporated, these extremes are reduced to 9.0% below and 4.4% above for January and February, respectively (see KPI - deseasonalized). What is even more disturbing is that the non-deseasonalized and deseasonalized KPI metrics provide distinctly different directional reads. In March, the non-seasonalized KPI actuals were a positive 6.7% versus a deseasonalized KPI metric of -1.3% - a total directional misread. This also occurs in August albeit at different levels, 9.8% versus -1.1%. In statistical parlance, this is known as a false positive and can cause management to make important business decisions based on misleading information. On the other side of the coin, there are three situations (April, May and September) where the KPI actuals indicate negative values (-6.0%, -4.1% and -6.5%) when in fact the deseasonalized KPIs are actually positive values (0.1%, 0.5% and 4.5%). These are known as false negatives and can once again severely mislead management concerning performance levels.

The ramifications here are quite serious with incorrect directional reads for five of the 12 months. Keep in mind that performance management implementations are only as effective as the data and KPI metrics that drive the tracking and measurement process. Management needs to stay focused on the real business issues, not on "pseudo" problems created by data and metric anomalies. In situations where senior executives are tracking KPIs to make important business and resource decisions, misleading KPI metrics can severely hinder the effectiveness and performance of the organization. The deseasonalization approach suggested here provides a methodology to eliminate the counterproductive activity of chasing down misleading KPI metrics and should be incorporated into the toolkit of all performance management practitioners.

Next month, we will extend the seasonality analysis approach to explore seasonality in context - how differences in seasonality indices can lead to improved understanding of the competitive marketplace.

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