Documents

4 pages
7 views

Hypothetical Testing

Please download to get full document.

View again

of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Description
Hypothetical Testing
Transcript
  49 Hypothesis Testing What is it? Hypothesis testing is a method or unearthing the key assumptions about what must be true about a strategic choice; it then seeks data to assess the likelihood that these assumptions are true. When do we use it? Hypothesis testing can be used to assess any strategic choice aced by a firm or analyst. It is a critical component o analyzing whether to pursue a particular course o action: whether to enter a new market, to invest in a research project, or to acquire another firm, or example. Why do we use it? Every strategic choice is a hypothesis —a proposition that a particular action will cause a determined outcome. Te conditions and assumptions the hypothesis rests on must be valid in order or it to be “true.” And so testing these assumptions is essential. Assumptions must be clearly articulated beore they can be tested. We have only two options or testing them: (1) use existing secondary data to conduct an analytical thought   experiment or (2) go out into the field and gather new data. Since finding new data tends to be costly and visible, we try to conduct early tests using data we already have. When strategies ail, it is always  because reality turns out to be different than we thought it would be: perhaps customers don’t want a new offering, a firm can’t execute it, partners don’t like it, or competitors copy it quickly. Hypothesis testing ocuses on identiying the make or break elements o a strategic choice. It involves taking a cold hard look at its key vulnerabilities. How do we use it? Step 1. Identify core assumptions. o get started, we first drill down to the core assumptions on which the success o a proposed stra-tegic action depends. In general, a common set o tests apply to just about any strategic decision:  THE STRATEGIST’S TOOLKIT50 5. Te Value est. I executed well, the strategic action will create value. For example, a new product offering will be desired by customers at a price that works. 6. Te Execution est. Te firm can create and deliver the strategic action at a cost that works. 7. Te Scale est. Te strategic action can be sustained and leveraged to create uture value that makes it worthwhile. 8. Te Deensibility est. Competitors can’t easily copy the strategic action.For example, imagine that you are a newly appointed business development officer at a home loan company recently acquired by a large credit card firm, and you have been given a mandate to iden-tiy and develop a new growth opportunity. Your new parent company is one o the largest credit card issuers in the United States with more than 60 million accounts worldwide and a reputation as a leader in direct marketing and online services. Besides credit cards, it offers a variety o finan-cial services such as auto loans, small business loans, home equity loans, and second mortgages. It does not, however, offer first mortgages. Being new to the mortgage business, or several months, you immerse yoursel in the ins and outs o the mortgage industry and finally identiy a number o what you believe to be high-potential growth opportunities. One o your avorite ideas is or a first mortgage product aimed at high-net-worth customers o the parent company, delivered with personal service through your bank branches. You believe that your proposed concept can pass all our o these tests: 1. Customers will value the convenience o personal attention rom your bankers, one-stop shopping or all their financial needs, and the superior rates your knowledge o their credit history will allow you to offer them. 2. Your parent company’s capability to deliver outstanding quality and ast turnaround time online or financial products is well known; you also have a set o bankers available already in your branches. 3. Historical data tells you that your parent organization has 35 million customers with excel-lent credit ratings; thus the uture potential business opportunity is substantial. 4. Already-established competitors in the marketplace, such as Bank o America and Wells Fargo, are presently distracted cleaning up the mess in their subprime mortgage business, and you see an opportunity to enter without ear o retaliation on their part.Tese are the oundation o your core assumptions or your hypothesis that the new mortgage product is worth pursuing. Make sure your assumptions relating to each individual test (value, execution, deensibility, and scalability) are as explicit as you can make them. Lay out specific  tests that your hypothesis must “pass” in order to move orward. Tese tests relate to your firm and your particular situation.Continuing with our example, recognize that these key assumptions revolve around a set o edu-cated guesses you’ve made about: Customers. Speciy why this concept will create superior value or customers beyond existing options, how much they will be willing to pay, and whether there are enough o them to constitute a market o sufficient size. In our example, as you think more deeply  HYPOTHESIS TESTING51 about your embedded assumptions, you realize that your value proposition rests on a number o important assumptions: one being that your high-net-worth customers value the ace-to-ace service delivered by your bankers. Your Organization.  How will the organization create and deliver the promised value, and what current capabilities will the organization leverage? You must also identiy which criti-cal capabilities are missing and whom you will partner with to obtain these. For example, while your parent organization’s superb online capability set is well established, your ability to deliver this new product also rests on the skill o your branch bankers. Do they have the necessary skill set, you wonder? You also realize that you are assuming that your parent’s scale in its Internet-driven credit card business will apply to scaling a personal service business. Competitors.  Which competitors are likely to be affected and how they will react? Tis would include assumptions about whether and why they are capable o copying the strategy quickly and how else they might interere with your ability to succeed. Tough competitors in the field are numerous, you are assuming that they will not notice or retaliate against your entry into their business.Determine which assumptions are most critical to the potential attractiveness o your proposed strategy. I you’ve been thorough, you have probably generated ar more assumptions than you can easibly test. Can you highlight the handul o assumptions that make or break your new idea? im-ing is also important to consider. Generally, the two tests that matter most in the early stages o a new idea are the value and execution tests. Scaling and deensibility come later, as you know more. Most new strategies ail the ormer (value test) rather than the latter (execution test). Reerring to our example, you may decide that the value test is the place to start. I a significant proportion o your target high-net-worth customers do not value either the personal service you intend to provide or the convenience o one-stop shopping, your concept is unlikely to succeed. Step 2. Conduct thought experiments. Next, identiy the data that would allow you to conclusively test the key assumptions. Having nar-rowed the assumptions down to a manageable number, you now begin to move rom unearthing assumptions to testing them. It is crucial here to think through what the data that would either confirm or disprove your hypotheses about the attractiveness o the new strategy look like. Here, you are identiying the inormation you need   and then figuring out how to get it. Sort the data you need into one o the ollowing three categories: what you know, what you don’t know and can’t, and what you don’t know but could. Let’s look at each category in turn: What you know.  Tese are the acts that you already have in your possession related to each assumption. Beware o belies masquerading as acts. Don’t conuse the two. Te doubters will help you with this by highlighting the areas where your personal (and sometimes optimistic) interpretations may be blinding you to some realities that need to be acknowledged. In our example, you know quite a bit. For instance, you know that your  THE STRATEGIST’S TOOLKIT52 parent company has a great capability set or doing financial transactions online and that it has a lot o high-net-worth customers already. What you don’t know and can’t know. Tis is the stuff that you can’t know without a crystal ball. It is the land o true uncertainty—the land o the unknowable. No number o experiments—either thought or real—can resolve this uncertainty. Te only thing you can do here is predict. For example, there are some macro issues that matter to your new concept’s success—issues such as housing starts and prime rate levels. You could make some predictions here, but you aren’t sure how to really test these, except against “expert” opinions. What you don’t know but could.  In any situation, there is a lot o stuff that is knowable—but you haven’t yet taken the time to go and get the data to know it yet. Generally, this can be an expensive proposition, and you don’t want to chase data you don’t need. Tat is why it is so critical to be hypothesis-driven in your approach: identiying only the really important data and then expending the effort to go get them. Some o these data may already exist elsewhere and you just need to go out and collect it. Tis is why consulting firms flour-ish—this is what they do. Going back to our example: As you reflect upon it, you realize that there are a lot o customers out there that you could identiy, already in the parent’s database, whose views about issues such as convenience and personal service you could better understand. Tere is also probably some inormation on the kind o qualifications the bankers in your branches have floating around somewhere, maybe in HR.Data in category #3 (what you don’t know but could) lends itsel to thought experiments. Identiy what it would take to get the data quickly. You are going to have to construct some data, which means not relying on what your internal accounting systems or industry trade group—or whom-ever—decides to give you. In our example, you decide that determining how your target customers eel about personal service and value does not lend itsel to a thought experiment. Afer all, they are already there in the database, you know who they are, and they are reasonably inexpensive to reach. Why conduct a lot o analyses based on past data that may not be reflective o what they really think, when you can easily just ask some o them?Design your thought experiment paying special attention to the data that could prove you wrong. Go collect data and begin to validate or invalidate the core assumptions underlying your assumptions. FURTHER READING Chapter 9 o Jeanne Liedtka and im Ogilvie  , Designing for Growth  (Columbia Business School Publishing, 2011).
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x