8. Why should I care?
“Those who fail to plan, plan to fail.”
9. Why should I care?
“Those who fail to plan, plan to fail.”
Financial forecasting helps
◉ Make better decisions
◉ Prioritize tasks
◉ Assess performance
◉ Valuate businesses
10. Why should I care?
“Those who fail to plan, plan to fail.”
Financial forecasting helps
◉ Make better decisions
◉ Prioritize tasks
◉ Assess performance
◉ Valuate businesses
11. Why should I care?
“Those who fail to plan, plan to fail.”
Financial forecasting helps
◉ Hire
12. Why should I care?
“Those who fail to plan, plan to fail.”
Financial forecasting helps
◉ Hire
&
◉ Fundraise
16. “
A financial forecast is an
economist's best guess of what
will happen to a company in
financial terms over a given time
period.
17. “
Using historical internal accounting and sales data,
in addition to external market and economic
indicators, a financial forecast is an economist's
best guess of what will happen to a company in
financial terms over a given time period.
18. “
Using historical internal accounting and sales data,
in addition to external market and economic
indicators, a financial forecast is an economist's
best guess of what will happen to a company in
financial terms over a given time period.
19. “
Using historical internal accounting and sales data,
in addition to external market and economic
indicators, a financial forecast is an economist's
best guess of what will happen to a company in
financial terms over a given time period.
22. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
24. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
25. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
26. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
28. Moving average: 3 month data
August September October November December
Revenue 800 1000 1000 ? ?
3 mo SMA - - - ? ?
800 + 1000 + 1000
3
= 933.33
29. Moving average: 3 month data
August September October November December
Revenue 800 1000 1000 ? ?
3 mo SMA - - - 933.33 ?
30. Moving average: 3 month data
August September October November December
Revenue 800 1000 1000 1050 ?
3 mo SMA - - - 933.33 ?
31. Moving average: 3 month data
August September October November December
Revenue 800 1000 1000 1050 ?
3 mo SMA - - - 933.33 ?
32. Moving average: 3 month data
August September October November December
Revenue 800 1000 1000 1050 ?
3 mo SMA - - - 933.33 1016.66
33. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
36. Exponential smoothing
August September October November December
Revenue 800 1000 1000 1050 ?
3 mo SMA - - - 933.33 1016.66
3 mo EMA 800 880 928 976.8 1006.08
37. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
38. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
40. Trend
Growing or not growing
Cyclicality
WordPress updates
Seasonality
Black Friday
41. Qualitative
◉ Customer research
◉ Market research
◉ Delphi method
Methods
Quantitative
◉ Time series
○ Rule of thumb
○ Smoothing
○ Decomposition
◉ Causal
○ Regression analysis
42. Y = a + bX + u
Linear regression
Dependent variable
Intercept
Coefficient
Independent variable
Residual
43. Y = a + bX + u
Linear regression
Thing you want to
know
Where
we’re
starting
How much
X matters
Thing you think will
influence Y
How good we feel
about this
relationship
44. Y = a + bX + u
Linear regression
Return on
WordCamp
attendance?
45. Y = a + bX + u
Linear regression
Return on
WordCamp
attendance?
Based on # of people
you networked with?
46. Y = 50 + 25X + u
Linear regression
Return on
WordCamp
attendance?
Based on # of people
you networked with?
Where
we’re
starting
How much
X matters
47. Y = 50 + 25X + 25
Linear regression
Return on
WordCamp
attendance?
Based on # of people
you networked with?
Where
we’re
starting
How much
X matters
How good we feel
about this
relationship
52. Process
1. Identify the problem
2. Identify relevant variables
3. Decide how to collect data
4. Make assumptions
5. Choose a model that fits
6. Forecast
7. Verify
55. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales
56. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales, # of tickets
57. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales, # of tickets,
minutes to resolve each ticket
58. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales, # of tickets,
minutes to resolve each ticket
3. Decide how to collect data: HelpScout
59. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales, # of tickets,
minutes to resolve each ticket
3. Decide how to collect data: HelpScout
4. Make assumptions
60. How much more support for an
increase in sales?
FACT: Last week, we sold 20 licenses. We also had 10
new support tickets. The week before we sold 20, and
had 12 new support tickets.
ASSUMPTION #1: We have 11 support tickets per
every 20 new sales.
61. How much more support for an
increase in sales?
FACT: Data collection in HelpScout says we spent an
average of 15 minutes on each ticket.
ASSUMPTION #2: Each ticket is 20 minutes of
support tech time.
62. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales, # of tickets,
minutes to resolve each ticket
3. Decide how to collect data: HelpScout
4. Make assumptions
5. Choose a model that fits
63. How much more support for an
increase in sales?
August September October November
# of new
sales 60 70 80 ?
# of addt’l
tickets 33 39 44 ?
Addt’l
minutes of
tech time
495 585 660 ?
64. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales, # of tickets,
minutes to resolve each ticket
3. Decide how to collect data: HelpScout
4. Make assumptions
5. Choose a model that fits
6. Forecast
7. Verify
65. How much more support for an
increase in sales?
August September October November
# of new
sales 60 70 80 90
# of addt’l
tickets 33 39 44 50
Addt’l
minutes of
tech time
495 585 660 750
66. How much more support for an
increase in sales?
1. Identify the problem
2. Identify relevant variables: # of sales, # of tickets,
minutes to resolve each ticket
3. Decide how to collect data: HelpScout
4. Make assumptions
5. Choose a model that fits
6. Forecast
7. Verify
67. How much more support for an
increase in sales?
August September October November
# of new
sales 60 70 80 100
# of addt’l
tickets 33 39 44 46
Addt’l
minutes of
tech time
495 585 660 690
69. Y1
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TOTAL
Revenue 0 0 0 0 0 0 500 600 1200 1300 1500 1500 6600
Costs 0 0 0 0 0 400 100 100 200 200 200 200 1700
Profit 0 0 0 0 0 (400) 400 500 1000 1100 1300 1300 4900
70. Y1
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TOTAL
Revenue 0 0 0 0 0 0 500 600 1200 1300 1500 1500 6600
Costs 0 0 0 0 0 400 100 100 200 200 200 200 1700
Profit 0 0 0 0 0 (400) 400 500 1000 1100 1300 1300 4900
71. Y2
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TOTAL
Revenue 1600 1700 1800 1900 2000 1500 1500 1400 2400 2500 2600 2700 6600
Costs
Profit
72. Y1
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TOTAL
Revenue 0 0 0 0 0 0 500 600 1200 1300 1500 1500 6600
Costs 0 0 0 0 0 400 100 100 200 200 200 200 1700
Profit 0 0 0 0 0 (400) 400 500 1000 1100 1300 1300 4900
73. Y2
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TOTAL
Revenue 1600 1700 1800 1900 2000 1500 1500 1400 2400 2500 2600 2700 23,600
Costs 200 200 200 200 200 200 200 200 400 400 400 400 3,200
Profit
74. Y2
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec TOTAL
Revenue 1600 1700 1800 1900 2000 1500 1500 1400 2400 2500 2600 2700 23,600
Costs 200 200 200 200 200 200 200 200 400 400 400 400 3,200
Profit 20,400