Transcript Day 1 - Hotel Profit Optimization

Science of Hotel Optimization
Rooms Revenue Workshop
Day 1: Data
Day 2: Analysis
Day 3: Optimization
50 minute periods.
10 minute break
every 50 minutes.
http://www.forsmarthotels.com/sohodocs
© Origin World Labs
SOHO Day 1
2
Ask every question.
Dozens of Books, Lectures, Courses in 4 hrs.
Contribute examples.
Give me a Pace.
I will go back.
© Origin World Labs
SOHO Day 1
3
Day 1 Objectives
My Notes
Hour 1
 The Analytic RM
 Hotel Data Science
 Databases & Data Access
Hour 2
 PMS Databases
 MS Query Tool
Hour 3
 Opera Data Tables
 OLAP Tools (Pivot Table)
 SQL Queries
Hour 4
 SQL Queries
 Excel Criteria Functions
© Origin World Labs
SOHO Day 1
4
What is Analytics?
My Notes
• Not Reporting.
• Misused and Abused.
e.g. “Luxury”
Applying mathematics to
data to extract insights that
lead to better decisions.
© Origin World Labs
SOHO Day 1
5
The ‘M’ Word
© Origin World Labs
SOHO Day 1
6
The Analytic RM
My Notes
• The Talent Shortage
• 50% shortage in analytic positions by
2018.
• The Hotel RM experience.
• Analytic maturity.
• 1-2% use RMS.
• Black box algorithms.
• Era of Rate Distribution ROI is over.
• Incremental value of RM.
© Origin World Labs
SOHO Day 1
7
Who would you pick to do your taxes?
Bookkeeper
© Origin World Labs
Accountant
SOHO Day 1
8
The Analytic RM
Clerical RM
Analytic RM
Rate Distribution
Analysis
Manual Reporting
Modeling Decisions
Comp Set focused
Patterns focused
Rooms Revenue
Total Profit
STR Report
Strategy ROI
My Notes
© Origin World Labs
SOHO Day 1
9
The Revenue Manager of 2023 Skillset
Analytics Toolbox
Data & Granularity
Drivers
Forecasts
Market
Biases
Probability
Statistics
Choice & Demand
Optimization
Modeling
Guest Behavior
Rhythm of Business
Communication
© Origin World Labs
SOHO Day 1
10
The Hotel Data Scientist
My Notes
• Harvard Business Review.
• Many disciplines.
• Broad skill set.
• Not from hospitality.
• Ability to think “nerd” and
speak “business”.
• One foot in IT and one foot in
the boardroom.
• Extracts profit from data.
• Start in RM.
“This Workshop is your launch point.”
© Origin World Labs
SOHO Day 1
11
Data Science Pyramid
SOHO
Data Science
Prescriptive
Optimization
Day 3
Predictive
Analysis &
Classification
Day 2
Descriptive
Data Extraction
Day 1
© Origin World Labs
SOHO Day 1
12
Your new world view
• Data in its native environment.
My Notes
• At the databases level
• Think “The Matrix”.
• Put your “tech” hat on.
© Origin World Labs
SOHO Day 1
13
Databases
• A set of spreadsheet-like tables.
• Multiple tables for efficiency.
• Works hidden behind transaction
software.
© Origin World Labs
SOHO Day 1
My Notes
14
Relational
Reservation Entry
Occupancy Graph
Client
Reservation
Guest Profile
Room Types
OOOs
Database
© Origin World Labs
SOHO Day 1
15
Simple PMS Schema
Unique keys connect each record(row).
Nightly
Reservation
Guest Profile
Date
Reserv ID
Guest ID
Rate
Arrival
Name
Reserv Date
Departure
Address
Source
Guest ID
Email
Reserv ID
Room Type
Phone
You must reconnect these tables to get all the data.
© Origin World Labs
SOHO Day 1
16
Databases: Important Definitions
My Notes
Table: related data grid.
Schema: Organization of tables
Fact : Transaction
e.g. Rate Paid one night
Dimension : Explains Fact
e.g. Source, Room Type
Granularity : Level of detail
e.g. POS has ticket detail
Columns: Field
Data Dictionary: Explains tables
& columns.
Warehouse: An organized copy.
© Origin World Labs
SOHO Day 1
17
PMS Database Vendors
Two licenses
© Origin World Labs
PMS
Database
Opera
Oracle
Agilysis
Microsoft
RoomMaster
Sybase
SOHO Day 1
18
Opera Tables: Reservations History
My Notes
Sample company
Upper Creek Resort
100 Rooms
RESERVATION_DAILY_ELEMENTS
Organized by Night
RESERVATION_DAILY_ELEMENTS_NAME
Organized by Reservation
© Origin World Labs
SOHO Day 1
19
ODBC
My Notes
• Open Database Connection.
• “Like rubbing the lamp.”
• Access data via ODBC client.
• Driver available from each
database vendor (not PMS) for
FREE.
• Usually installed by IT.
• Oracle ODBC client at
Instructions at sohodocs.
© Origin World Labs
SOHO Day 1
20
Four Data Table Analysis tools
1. Query Wizard – MS Query
2. OLAP – Picot Table
3. SQL
4. Excel – Multi-Criterion Functions
© Origin World Labs
SOHO Day 1
21
MSQUERY
• The most common operation for a
business analyst is the query.
My Notes
• MSQuery is the simplest way to query
a database.
• Built into Excel.
• Database agnostic. Vendor Neutral.
• Creates a Real-time link to data.
Data > From Other Sources > From Microsoft Query
© Origin World Labs
SOHO Day 1
22
MSQUERY
My Notes
Connect
Update
Refresh
© Origin World Labs
SOHO Day 1
23
Query Wizard
Select Table
Filter
© Origin World Labs
Select Columns
Sort
SOHO Day 1
Return
24
MS Query power tools
My Notes
• Criteria > Add Criteria
• Table > Add Tables
• Drag and Drop Join
© Origin World Labs
SOHO Day 1
25
Excel Table
My Notes
•
•
•
•
•
•
•
•
•
Behaves like a database table.
For related data.
Header Row
Filters
Sorting
Calculated Columns
Formatting
Insert/Delete
Structured references
Instead of C2
[@[RESV_DAILY_EL_SEQ]]
© Origin World Labs
SOHO Day 1
26
Pivot Table
My Notes
• OLAP Tool
• Fast way to investigate data.
• Multi-dimensional
• Multi-perspectives
• Drill-down
• Slicing
more at sohodocs
Insert > Pivot Table
© Origin World Labs
SOHO Day 1
27
Pivot Table
My Notes
What you are measuring
• Values
The Dimensions
• Columns
• Rows
Slice of the data
• Filters
Tips
• Start with the end in mind
• Creative Thinking
© Origin World Labs
SOHO Day 1
28
SQL: Standard Query Language
• Common Language for Database
My Notes
• A very specific way to ask a
database a question.
• Universal ANSI standard.
• We will use the query part.
• DML Data manipulation language.
• “The coding” is sensitive.
© Origin World Labs
SOHO Day 1
29
Types of SQL
My Notes
Developed to connect data source to a programming language.
Slight variations in code.
PMS
Database
SQL
Opera
Oracle
PL/SQL
Agilysis
Microsoft
T-SQL
RoomMaster
Sybase
T-SQL
© Origin World Labs
SOHO Day 1
30
SELECT
My Notes
• To select the columns to return.
• What is to be returned.
• “*” represents all
SELECT column_name,column_name
© Origin World Labs
SOHO Day 1
31
FROM
My Notes
• Defines the Table(s) to be
used
• Always required.
[Database].[Table]
SELECT column_name,column_name
FROM table_name;
SELECT
FROM
© Origin World Labs
RESV_NAME_ID
`C:\SOHODAY1.xlsx`.`Element$`
SOHO Day 1
32
WHERE
My Notes
• Your filters
• Use operators
– =, >, >=, <, <=
– Between
– In (list)
• Multiple Conditions joined by “and”
SELECT column_name,column_name
FROM table_name
WHERE column_name operator value;
SELECT *
FROM `C:\SOHODAY1.xlsx`.`Element$`
WHERE (RESERVATION_DATE>{ts '2012-01-01 00:00:00'})
© Origin World Labs
SOHO Day 1
33
JOINS
SELECT
FROM
*
`C:\SOHODAY1.xlsx`.`Element_Name$`
`C:\SOHODAY1.xlsx`.`Elements$`
WHERE `Elements$`.RESV_DAILY_EL_SEQ =
`Element_Name$`.RESV_DAILY_EL_SEQ
My Notes
© Origin World Labs
SOHO Day 1
34
GROUP BY
My Notes
• Group Functions
• Common Arithmetic
Count, Sum, Max, Min, Avg
SELECT
RESERVATION_DATE,
SUM(RATE_AMOUNT),
SUM(QUANTITY)
FROM `C:\SOHODAY1.xlsx`.`Element$`
WHERE (RESERVATION_DATE>{ts '2012-01-01 00:00:00'})
GROUP BY RESERVATION_DATE
© Origin World Labs
SOHO Day 1
35
ALIASES
My Notes
• Renames a column in query
• Careful not to use database or
SQL words
SELECT
RESERVATION_DATE STAY,
SUM(RATE_AMOUNT) REVENUE,
SUM(QUANTITY) NIGHTS
FROM `C:\SOHODAY1.xlsx`.`Element$`
WHERE (RESERVATION_DATE>{ts '2012-01-01 00:00:00'})
GROUP BY RESERVATION_DATE
© Origin World Labs
SOHO Day 1
36
Excel Vlookup
• Go get data according to a match
• Searches first column.
• VLOOKUP( DATA TO MATCH, DATA GRID, COLUMN TO
SEARCH, TYPE OF MATCH)
My Notes
© Origin World Labs
SOHO Day 1
37
Excel Multiple Criteria Functions
My Notes
• COUNTIFS
–
COUNTIFS(Column, Criteria,Column2, Criteria2,…)
• SUMIFS
–
SUMIFS(Column to Sum,Column,Criteria1,…)
• AVERAGEIFS
Criteria filter – “operator”&
e.g. “>=“&25
© Origin World Labs
SOHO Day 1
38
What we covered?
My Notes
 The New RM
 Data
 Databases
 Tables
 Columns
 Dimensions
 MSQuery
 Pivot Table
 SQL
 Excel
© Origin World Labs
SOHO Day 1
39
Next Friday: Analysis
 Probability
 Variance
 Expected Value
 Demand Curves
 Forecasting
© Origin World Labs
SOHO Day 1
40