Three Critical Steps for Extracting Great Insights (Step 1: Identify Parameters)

This post covers this element(s) of GAMMA PI: #1. Generate Informed Insights

Think of your favorite CNBC journalist and ask why he or she is likely to get more information from a CEO than a typical analyst. If you don’t have an answer, it’s because the top journalists have been trained to use best practices for interviewing (or even “interrogating”) which can make them incredibly effective.

Throughout my career as an equity research analyst, I observed some analysts were much better than others at extracting insights from information sources (e.g. proprietary industry sources, company management, etc.), but I didn’t know exactly why. Now in the role of training analysts, I’m routinely asked, “How do I get more insights from others?” In my effort to answer this question and determine why some analysts are better than others in getting insights from others, I identified the best practices in this area, which primarily come from journalism, the legal industry and law enforcement (yes, some of the most sophisticated interviewing practices are used by law enforcement).

You’ll get a better answer if you craft a question to include a specific parameter

I put all of these best practices into a fairly simple 3-dimension model, which I’ve dubbed “ICE™”:

ICE Framework

Due to space constraints, I’ll delve into the “I” in this post and then move onto “C” and “E” in future posts. These tactics can help in almost any exchange for information, ranging from high profile one-on-one interviews to something as simple as an email exchange.

Identify Parameters

Which of these questions posed to company management is more likely to evoke an answer useful for analyzing a stock:

  1. “Is there anything we should discuss in your labor expense line item?”
  2. “Was the 3% wage inflation in the recent quarter, which compares to only a 1% increase for the prior four quarters, because the company is beginning to lap its productivity initiatives from last year?”

Here’s the simple fact: If your question (or lead-in to your question) references specific parameters (e.g. historical trends, consensus expectations, your forecast, etc.) you have a much higher likelihood of getting an insightful answer that will help in stock picking.

Hopefully this underscores that the more you know before the interview, the more you can refine the question (and thus the response) to get to the heart of the issue. The problem is you don’t have endless hours to prepare for each interview, which is why it’s a best practice to focus only on the 1-4 critical factors for a stock. Other than building rapport with the interviewee, all of your questions should help to clarify assumptions around a stock’s critical factors. (This is also the first step of our ASPIRE™ framework for generating unique insights.)

In Exhibit 1 below we provide questioning strategies that reference specific parameters, which should result in more specific answers from the interviewee. We have designed the table with the most effective strategies on the top, but since you won’t always have public goals from the interviewee or a third-party forecast, you may need to move down to other strategies (which are still more effective than questions lacking parameters).

Exhibit 1: Questioning Strategies That Reference Specific Parameters

Questioning Strategies That Specify ParametersExamples
Reference Interviewee’s' Stated Goals (when available): Ensures response is reconciled with interviewee’s goalsCan you help me get comfortable with management's long-term goal of 1% labor cost inflation, given that it was up almost 3% in the recent quarter?
Reference Credible Third-party Forecast (when available): Keeps interviewee within the parameters of this credible forecastDo you believe the Global Aircraft Association’'s long-term forecast for 6% intra-Asia passenger traffic is realistic?
Reference Past or Continuation of Trends: Frames the past which helps to keep interviewee’s response more realisticShould I assume that the lack of labor productivity improvement in the most recent quarter will continue?
Offer Options or Scaling: Keeps interviewee’s' response within a predefined rangeOver the next 18 months, is it more realistic that labor inflation will return to the highs of 6%-7%, as seen 4-5 years ago, or potentially be flat, similar to a target set by your largest competitor?
Hypothesize: Challenges interviewees to respond with their own hypothesisI forecast the major productivity improvements of the prior year will slow to less than 1%, given the weak numbers in the most recent quarter.
Make Implications: Uses your insights to build a well-constructed implication that will likely elicit a quantifiable responseYour largest competitor has stated one of its competitors has been dropping pricing and we see in your most recent quarter, your average selling price was down. Can I assume you are the company they’'re referencing?
Triangulate: Uses information from other related areas to narrow possible outcomesI’ve heard management state it generates $60 per unit from product sold in China. If your company sells 50,000 units per day in Shanghai and we’ve been told Shanghai is over 50% of your China revenue, the math suggests you generate $1.4 billion per year in China. Are there any elements of my assumptions that appear way off?

Give the Interviewee Some Room…But Not Too Much

The goal for using the “options”, “scaling” and “hypothesize” question types above is to allow the respondent some room to answer the question in a manner that makes them feel comfortable, while not avoiding the question altogether. Notice in the second example below, the analyst is making the assertion that the employees will unionize and so if the interviewee responds, he or she is not stating the employees are going to unionize:

  • Avoid being this direct: “Are your California employees likely to unionize?”
  • Instead, hypothesize: “I suspect your costs will increase if the unions are successful in unionizing your California operations.”

Technically, the second bullet above isn’t a question, but the interviewee will likely respond with their input.

Great managers thrive on being great “problem solvers” and so, by their very nature, they can’t allow a problem (your incorrect range) to go by without being solved. I’ve found interviewees more likely to answer my questions if I provide “response scaling” such as: “Do you expect to add 30 or 35 planes to the fleet?” (Notice the presumptuous question type) rather than, “How many planes do you expect to add?” If it turns out your 30-35 range is way off, management will usually try to correct. In another example, “How many days this quarter did you not meet your revenue target?” will not likely get an answer whereas, “Did you miss your target on at least 20 days this quarter?” gets under management’s skin because it doesn’t want to be portrayed in the wrong light, which means it will be motivated to provide the right answer.

You may even throw out an absurd hypothesis so that management will put parameters around the real answer. If you suspect margins in Germany are slightly better than those in France, you could say, “I suspect your margins in Germany are twice as high as in France.” But don’t do this too often or the interviewee will see through it and stop providing insight.

Rather than seek a specific number, ask for an estimate or range of estimates:

  • Avoid being this direct: “What is your R&D spending going to be next year?”
  • Instead, ask for a range: “Can you give us some thoughts about the higher and lower ends of R&D spending likely for next year?”

Ask what could go wrong in achieving a parameter:

  • Avoid being this direct: “Will you achieve your cost-cutting target this quarter?”
  • Instead, ask what would cause a target from being achieved: “What would prevent your company from achieving its cost-cutting target this quarter?”

Conclusion (for now)

There’s more to cover in the ICE™ framework, which I’ll need to discuss in a future post. I hope I’m conveying that if you identify parameters before your interview or email exchange, you’re likely to get a better response. If you’d like to learn how to apply this framework, AnalystSolutions offers an on-demand workshop that delves deeper: Generate Differentiated Insights Through Better Discovery, Questioning and Influencing.

This Best Practices Bulletin™ targets activity #1, “Generate Informed Insights” within our GAMMA PI™ framework.

Visit our new Resource Center to find more helpful articles, reference cards, and advice towards your growth as an Equity Research Analyst.

©AnalystSolutions LLP All rights reserved. James J. Valentine, CFA is author of Best Practices for Equity Research Analysts, founder of AnalystSolutions and was a top-ranked equity research analyst for ten consecutive years

AnalystSolutions QRC, Information Sources Pros and Cons v2.7P

Sign Up for Our Reference Card of the Month Club

We will send you a reference card each month, highlighting a key best practice critical to success

Let us know your interests

AnalystSolutions respects your privacy and will never share your information with others. You can remove yourself from any distribution with one click.

Assessment and Book Samples:
Workshop Free Previews:

You have Successfully Subscribed!

Let us know your interests

AnalystSolutions respects your privacy and will never share your information with others. You can remove yourself from any distribution with one click.

Assessment and Book Samples:
Workshop Free Previews:

You have Successfully Subscribed!

Evalulate our tools for free

We offer free full versions of our content for evaluation purposes (valued at over $700) to professionals holding the following roles: 

  • Manager of buy-side or sell-side analysts (DOR or PM)
  • CFA Society leader responsible for programming
  • Professor teaching students pursuing careers in equity research

You will receive:

  1. Free GAMMA PI™ assessment
  2. Free access to our presentation "Are You a Great Analyst?"
  3. Free full workshop (not "Preview" version), including all materials (Learner Workbook, Slides and Quick Reference Cards)
  4. Sample chapter of Best Practices for Equity Research Analysts

To ensure a quick verification of your role above (usually within one business day), please provide the email address issued by your organization.


You have Successfully Subscribed!

Try 2 of the 15 GAMMA PI™ Modules for Free

You will be directed to the FREE SAMPLE EDITION of our GAMMA PI™ Equity Research Analyst Self-Assessment.  Please provide a valid email address to receive  your results (your email address remains confidential within AnalystSolutions).

You have Successfully Subscribed!