HypothesisLibrary
 Share
The version of the browser you are using is no longer supported. Please upgrade to a supported browser.Dismiss

 
View only
 
 
ABCDEFGHIJKLMNOPQRST
1
Hypothesis Library
2
Overview and Instructions
3
4
Overview
5
6
This spreadsheet is a template for capturing, assessing, prioritizing, and tracking hypotheses over time. It is structured around the "A.D.A.P.T. to Act and Learn" framework:
7
8
Align...on goals and KPIs (not included in this template)
9
10
Discover...hypotheses across the company. Capture these in a format of "I believe…[hypothesis]" and "If I'm right…[action]." This construct is intended to make hypothesis development as accessible as possible to as many people in the business as possible. Good hypotheses can come from anywhere!
11
12
Assess...the hypotheses and your approach (because hypotheses can be tested in many ways beyond simply A/B or multivariate testing)
13
14
Prioritize...the hypotheses -- guided by the various factors that make prioritization messy
15
16
Test...the highest priority hypotheses
17
18
Act...on the results -- report the results and then track to ensure action is taken
19
20
Learn...for the future -- capture and apply deeper learning that can be applied across the organization
21
22
Every organization is different, so this template is intended to be modified to meet the unique needs of your organization.
23
24
How to Use This Template
25
26
The meat of this template is the Hypothesis Library worksheet. While the worksheet contains a number of columns, they are organized such that groups of columns can readily be hidden depending on what aspect of the process is being worked on at the time:
27
28
Discover
29
Any new hypothesis that gets generated (ideally, hypotheses can be entered at any time, whenever someone has an idea that might warrant exploration; this can be supplemented with formal hypothesis-generation exercises that occur on a fixed schedule) should be entered as a new row in the table in the "I believe..."/"If I'm right..." format.
30
31
Assess
32
On a regular schedule, new hypotheses should be assessed to determine what the best approach would be to test the hypothesis. This is a key input into the prioritization -- both assessing the effort required to test the hypothesis, as well as to factor in practical considerations about available resources to implement the approach. The available options for the approach can be adjusted on the Settings and Lookups worksheet
33
34
Prioritize
35
The spreadsheet includes suggested criteria to factor into the prioritization of the hypothesis. These factors should be quickly assessed and are intentionally simple High/Medium/Low assessments. While the selection for each criteria contributes to the prioritization score, the score is not intended to be an absolute judgment on the prioritization. Given a long list of hypotheses, it is intended to surface those that are most promising and suppress the ones that are least promising. By default, these criteria are not evenly weighted, and the weighting of both each criteria and each value within each criteria can be adjusted on the Settings and Lookups worksheet.
36
37
Test
38
This workbook is not intended to track all of the details of each hypothesis. However, the Status column can be used for high-level tracking. The available values for the Status can be adjusted on the Settings and Lookups worksheet. In addition, the worksheet includes columns to capture the date results of the hypothesis test were delivered, as well as both a quick summary of the result and a field to capture the location of more detailed results.
39
40
Act
41
Just because an analysis returns actionable results does not mean that action will actually get taken ("Curses, stakeholders! Curses!!!). By tracking the date that action is taken, the analyst takes responsibility beyond simply "delivering the results" and is actually responsible for following up to ensure that the results led to a decision and action.
42
43
Learn
44
The world would be just swell if every analysis -- every hypothesis tested -- led to deep and insightful learning. The world would also be really swell if mattresses never grew old, Butterfingers were healthy, and Firefly returned for a 10-season run. Alas! 'tis not so. BUT, every hypothesis tested should be followed by the question, "Did we gain any deeper learning beyond just this focused hypothesis?" If so, it should be captured, and this worksheet provides a way to do so.
45
46
Feedback and Questions
47
48
If you have feedback or questions about this spreadsheet, contact Tim Wilson at tim.wilson@webanalyticsdemystified.com.
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Loading...
 
 
 
Overview and Instructions
Hypothesis Library
Settings and Lookups