Contents:

1. What and how
2. Data file missing
3. OTB alert
4. Forecast alert
5. What is an outlier?
6. Threshold calculations

What and how?

The Actionable insights highlights points to review before making changes to the Live forecast. Several alerts can appear. Below are a few examples.

1. This example informs you of a normal and/or priority alert (always in red text). Its triggered based on a threshold calculation. Click Review and take action! to get started.

2. This example shows “You have no required actions for now”.

There are 3 types of alerts can appear on this tile. Missing data files, OTB (on the book) values alerts, and Forecast alerts. Alerts arise from the latest imported data updated each morning. Missing files alert appears irrelevant of the period chosen, however the OTB and Forecast alerts is directly impacted by the period chosen.

The threshold calculations for these alerts based on quartiles. Its a statistical term used to describe a data set into four equal parts, based on the values and how they compare to the entire data set. There are three quartiles, first, second and third represented by Q 1, Q 2 and Q 3. Q 2 is nothing but the median, since it indicates the position of the item in the list and is a positional average.

Workflow of alerts

Data files missing

1. There are 3 types, OTB, revenue and rooms/guests. Below is an example of an OTB missing file. This is a “red flag”, missing file equals incorrect figures. Deal with it as soon as possible i.e. follow up with your IT or person responsible. The alert is removed from the system once said file is imported into PMI.

• Click Review and take action! to get started

OTB alert

1. There are 3 types, Room nights, ARR and guest nights. Separately checked and independent of each other. These are noticeable changes, based on the PMI Prediction. I.e. the current OTB for room nights and ARR are compared to the historical OTB values for the same season and weekday, and with the same lead time.

• Room nights thresholds: A normal alert is triggered if the value is 25% higher than the higher (third) quartile, or lower than 25% of the lower (first) quartile. It becomes a priority alert if the value is 50% higher than the higher (third) quartile, or lower than 50% of the lower (first) quartile.
• ARR thresholds: A normal alert is triggered if the value is 10% higher than the higher (third) quartile, or lower than 10% of the lower (first) quartile. It becomes a priority alert if the value is 33% higher than the higher (third) quartile, or lower than 33% of the lower (first) quartile.
• Guest nights OTB guest factor for each day is compared to the Guest Night factor table (which is in Tools). Lead time is not considered for guest nights, as the OTB is compared to the season and weekday actual value. Please note that days with less than 20 room nights are not checked.
• Guest night thresholds: A normal alert is triggered if the factor is 50% higher/lower than the average. It becomes a priority alert if the factor is more than twice the normal average, or, lower than 1.00. In most cases 50% lower will bring the factor below 1.0, making it a priority alert.

1. Click Review and take action! or Show more depending where the alert appears.
2. Overview of all alerts for the period chosen appears. Confirm status of each alert by
3. Set status to This is correct, or Snooze.
4. This is correct – you agree with the said figures.
5. Snooze – you need more information, or awaiting feedback on a specific date. So if nothing changes in your PMS then it will reappear as an alert after the next import (which is usually the next day).
6. Click on pop out icon to view and it takes you directly to the table view to the specific day the alert is. If you want to make a change, click on the auto icon next to the date, a manual icon now appears and you can make your changes.
7. Click Save button on the popup to confirm all changes. A message appears to show that the graph and table is updated.
8. Click Refresh Insights to refresh the popup or Close when you have completed some tasks, and you will go back at a later stage to complete the rest of the period.

Snooze functionality

By using the snooze functionality, you need more information, or awaiting feedback on a specific date(s). So, if nothing changes in your PMS then it will reappear as an alert after the next import (which is usually the next day).

Forecast alert

1. There are 2 types Rates and Rooms.

• Room nights thresholds: A normal alert is triggered if the Live forecast is 25% higher or lower than the PMI prediction, and at least 20 rooms. It becomes a priority alert if the Live forecast is 40% higher or lower than the PMI prediction, and at least 30 rooms.
• ARR thresholds: A normal alert is triggered if the value is 10% higher than the higher (third) quartile, or lower than 10% of the lower (first) quartile. It becomes a priority alert if the value is 33% higher than the higher (third) quartile, or lower than 33% of the lower (first) quartile.

1. Click Review and take action!
2. Overview of all alerts for the period chosen appears. Confirm status of each alert by
3. Changing status to This is correct, Use PMI Prediction or Snooze
4. Click on pop out icon to view and it takes you directly to the table view to the specific day the alert is. If you want to make a change, click on the auto icon next to the date, a manual icon now appears and you can make your changes.
5. This is correct – you agree with the said figures.
6. Use PMI Prediction, and these figures will populate throughout the chosen period.
7. Snooze – you need more information, or awaiting feedback on a specific date. So if nothing changes in your PMS then it will reappear as an alert after the next import (which is usually the next day).
8. Click Save button on the popup to confirm all changes. A message appears to show that the graph and table is updated.
9. Click Refresh Insights to refresh the popup or Close when you have completed some tasks, and you will go back at a later stage to complete the rest of the period.

What is an outlier?

An outlier is something that is abnormal for the specific weekday and season. For example, If a property usually closes during Christmas, it is normal that these dates have 0 revenue. If one day then has sales, that day will become an outlier as 0 is considered as normal. 

A date set as an outlier will be disregarded when it comes to forecasting.  

Most outliers are detected automatically.  

Subject to user rights, you can also add manual outliers. To do this, go to Live forecast >Tools > PMI Live forecast > Tools > Seasons, and add a new label with an appropriate name. In the dropdown select ‘Outliers’.

Threshold calculations