1 Alpha Construction Group Housing Components Products Demand Forecast New Developments in Measurement and Analytics
1
Alpha Construction Group
Housing Components Products Demand ForecastNew Developments in Measurement and Analytics
2
Forward
• Many businesses are affected by the cyclical effects of
housing and remodeling demand, yet they have not
developed formal tools to allow these data to drive their
business plans and forecasts.
• Global-Analytics-Partners has significant experience in
this and can provide clients with an interactive tool which
will help them more accurately forecast and plan their
business performance.
3
Content
• Forecast Model Description
• Sales Predictions v. Actuals
• Forecast Summary and Assumptions for 2015
• Drivers of Growth for 2014
• Model Driver Relative Importance
• Model Driver Impacts and Sensitivities– Recurring Seasonality
– Housing Starts
– Remodeling Spending
– Order Backlog
– Price to Customer
• Forecast-Simulator Demonstration
• Summary
4
• The forecast model covers monthly periods from Jan 2009 to present. Four positive
drivers and one negative driver were used.
• The positive drivers include recurring seasonality, housing starts, Home Remodeling
Spend & Order Backlog. The negative driver was average price to the customer.
Home
Remodeling
Spend
Order
Backlog
Price-to-
Customer
Monthly
Unit
Demand
Housing
Starts
Recurring
Seasonality
5
-
20,00,000
40,00,000
60,00,000
80,00,000
100,00,000
120,00,000
Jan
.2006
May.2
006
Se
p.2
006
Jan
.2007
May.2
007
Sep
.2007
Jan
.2008
May.2
008
Se
p.2
008
Jan
.2009
May.2
009
Se
p.2
009
Jan
.2010
May.2
010
Sep
.2010
Jan
.2011
May.2
011
Se
p.2
011
Jan
.2012
May.2
012
Se
p.2
012
Jan
.2013
May.2
013
Sep
.2013
Jan
.2014
May.2
014
Se
p.2
014
Jan
.2015
May.2
015
Se
p.2
015
Forecast/Simulated Sales
Actual Sales
Forecast & Actual Sales Plot
Model R2=93.2%, Mean Absoluter Error +/- 4.1%
From the bottom of the recession in 2009, Components have been growing at a +9.5% annual rate per year.
6
Forecast Summary & Assumptions
• With modest but slightly improving growth in key drivers,
plus flat pricing, we would expect 2015 components
demand to improve to about +7.2% in 2015.
• This would generate near-term record components sales
of over 116 MM.
7
Sources of Growth
-1.0%
-0.4%
-0.04%
2.0%
2.2%
-1.5% -1.0% -0.5% 0.0% 0.5% 1.0% 1.5% 2.0% 2.5%
Price to Cust
Drawings
Repairs
Housing Starts
Baseline Momentum
Growth Impact
Growth Impact
• Based on the model, here is the contribution to Housing Components overall annual growth of +2.8%
8Relative Impact Importance of Housing
Component Business Drivers
33%
6%
25%
1%
35%
Importance of Sales Impact
Recurring Seasonality
Price to Customer
Housing Starts
Order Backlogs
Remodel Spend
Remodel Spend, followed by Seasonality and Housing Starts comprise the dominant
drivers of components’ demand.
9
Impact of Recurring Seasonality
8784
88
99 101
109 111114 113 111
98
83
40,00,000
45,00,000
50,00,000
55,00,000
60,00,000
65,00,000
70,00,000
75,00,000
80,00,000
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
Average Sales Impact of Recurring Seasonality
The “high season” for component demand is June-October
10
Price-to-Customer Demand Sensitivity
86,00,000
87,00,000
88,00,000
89,00,000
90,00,000
91,00,000
92,00,000
93,00,000
94,00,000
-6.0% -4.0% -2.0% 0.0% 2.0% 4.0% 6.0%
Mo
nth
ly D
em
an
d
Change in Price
Monthly Demand Impact from Change in Price to Customer
• The sensitivity of price-to-customer is slightly “inelastic” at -0.6, meaning a 1% increase would yield about a -0.6% decline in sales.
• This sensitivity indicates that price increases are likely to be modestly profitable. On average there is also about a 3 month lag between the largest change in sales and the change in price.
11
Housing-Starts Demand Sensitivity
50,00,000
60,00,000
70,00,000
80,00,000
90,00,000
100,00,000
110,00,000
-60.0% -40.0% -20.0% 0.0% 20.0% 40.0% 60.0%
Mo
nth
ly D
em
an
d
Change in Housing Starts
Monthly Demand Impact from Change in Housing Starts
• Housing starts can vary significantly from period to period. Overall, every 10% increase in starts generates about a +2.8% increase in Components sales.
• On average there is also about a 3 month lag between the largest change in sales and the change in housing-starts.
12
Remodeling Spend Demand Sensitivity
50,00,000
60,00,000
70,00,000
80,00,000
90,00,000
100,00,000
110,00,000
-60.0% -40.0% -20.0% 0.0% 20.0% 40.0% 60.0%
Mo
nth
ly D
em
an
d
Change in Drawings
Monthly Demand Impact from Change in Remodel Spend
• Remodeling Spending by Home Owners has the largest single impact of any driver. Every 10% change is connected to a 4.6% increase in component sales.
• On average there is also about a 3 month lag between the largest change in sales and the change in remodeling spending
13
Monthly Order Backlog Demand Sensitivity
90,00,000
90,10,000
90,20,000
90,30,000
90,40,000
90,50,000
90,60,000
90,70,000
90,80,000
90,90,000
91,00,000
-60.0% -40.0% -20.0% 0.0% 20.0% 40.0% 60.0%
Mo
nth
ly D
em
an
d
Change in Repairs
Monthly Demand Impact from Change in Order-Backlogs
• Backlogs have the least impact of any driver.
• There are also no lead-lag relationships between demand and
changes in repairs
14
Forecast-Simulation Tool
• Below is a rendering of the interactive-forecasting tool
which will be demonstrated
15
Summary
• Across 5 key business drivers, we have derived a model which connects to monthly Components demand with a 93% fit.
• For 2014, prior sales momentum and housing starts are the two leading positive drivers of annual +3.1% sales gain.
• With “modest” assumptions about the growth in key drivers, and flat pricing, for 2015, we expect to see demand growth of about +7.2% versus about +3.1% growth in 2014.
• Drawings, recurring seasonality and housing starts form the most dominant drivers of components monthly demand
16
Bangalore, IN Office:No. 141, 2nd Cross, 2nd Main,Domlur, 2nd Stage, Bangalore 560071Phone: +91 80 40917572, +91 80 [email protected]
Contact Us US Office:
Suite 100, 1780 Chadds Lake Dr, NE
Marietta, Georgia, 30068-1608
Atlanta, USA