Contradicting feedback on analysing graphs

Hi there, 

Sorry, but none of these is right. And I'm not sure that guy you spoke with is a McK Partner if that's what he advocated for. 

I'm always fascinated how candidates want to interpret an exhibit on the fly, when as a consultant I've been in a room with senior Partners with 20y of experience and they also went silent for a half a minute to just take in the exhibit. Nobody expects you to be a robot - it's ok to take time to think. Especially when you have little to no experience.

So how should you do it instead?

1. Read the exhibit with the interviewer. Read it, not interpret it. Use this as a chance to get a grasp of what the exhibit stands for, what are the types of data it contains and ask any clarifying questions you might have.

2. Take time. Not a lot. Somewhere between 20-60 seconds is fine. It's always better to take more time than have a shallow, non-insightful answer.

3. Get back to the interviewer with a structured answer. Focus ideally on at least 3 insights. For each of these insights, pick on a data point, explain what it means and then explain what the client should do as a consequence of this findings (this is what actually makes it an insight). 

4. For a distinctive interpretation, after you went through these insights, take a step back and try to paint the story for the client of how your refined hypothesis looks like now that you have all this additional data.

I run focused session with candidates on math and chart interpretation, so happy to discuss this with you. My candidates receive distinctive feedback from interviewers in terms of how they approach numericals and that's because they have the right technique.

Best,

Cristian

Hi there,

If you need more time, you should always take it.  I continuously repeat this to everyone that asks.  

Of course it's better to be quick, but it's most important to be right.  So these two recommendations actually converge, since you should try to do it in 10 seconds, but if you need 2 minutes, you just have to take it.

Also, there are a few techniques to think out loud, such as reading axes or calling attention to a few outstanding data points that can help you.

Hope this helps.  Please reach out for more.

Allen

Hi there,

Q: Which one of these alternatives makes the most sense?

You don’t necessarily need to take either 10 seconds or 2 minutes when you read a graph. Take the time you need to understand the graph, which depends on (i) your understanding of it and (ii) the graph itself. 

If you need 2 minutes, better to take the 2 minutes. The alternative is that you start talking without knowing what the graph is about.

Below are the steps I would recommend for graph analysis.

1. READ THE GRAPH

• Ask for some time to understand the graph. Usually 30 seconds are enough, but as mentioned above there is no hard rule.
• Summarize the graph. Read in particular the graph title, the legend, and the footnote. Candidates quite often skip this part and then make mistakes.

2. ANALYZE THE GRAPH AND FIND THE KEY INSIGHTS

• Repeat the question. Some candidates don’t repeat the question and then answer the wrong one. If you repeat the question, you can be sure you will answer what was asked.
• Provide an analysis related to the question. Identify the key insights of the graph based on the question. This is the most difficult part of graph analysis to master, as it is different in every graph. You can improve it with drills on your own or with coaching.

3. PROPOSE THE NEXT STEPS

• State your hypothesis or suggestion on what to do next. Present what should be done next to help further the client to reach the goal.
• Ask a question/propose an analysis related to what is needed to move forward. This will show you are able to drive the case forward.

Best,

Francesco

Hi there,

I think this is an interesting question that may be relevant for many people. I would be happy to share my thoughts on it:

• From my perspective, there is no one right answer. The key is to take the time you need to properly analyze the chart and come up with relevant insights.
• Moreover, it seems that receiving help from a coach might help you, especially given the conflicting information you've received so far. Analyzing different types of charts and graphs with a coach who can give you feedback on your approach and help you find the best way to analyze them might give you the confidence you need at the moment.

If you would like a more detailed discussion on how to address your specific situation, please don't hesitate to contact me directly.

Best,

Hagen

Agreed that being right is more important than being quick. Just think - in the final room when they're talking about applicants who are successful, will anyone say “they were quick with the graph but no real insights” or “they totally aced the graph” … the latter. For sure. 

With that being said, awkward silence can be real and nerve-wrecking. So I am a BIG fan of thinking out loud. Some things to mention:

1. What you're looking at (is this a chart over time or with stagnant data)

2. Any hypothesis about what you THINK it will show (we love a good hypothesis) 

3. What you're going to look at first, then second, etc - telling me what you're about to do makes me feel confident you're on the right path…and if you're not I can correct it before you've gone too far

Hi there,

Both inputs are very absolute and aren't very helpful.

Here's an example of where the Partner is technically wrong: McKinsey cases don't generally conclude with a final recommendation and that's just something that he/she chose to do, but it's not by the book. Problem is that Partners set the rules of the game, even if it's not by the book. That's why I recommend dedicated coaching for Partner rounds.

As for graphs, I generally recommend the following:

• 1st step: You should ask for and take the time you need to feel comfortable to talk about the obvious insights (1st level analysis) but expect to get a “how's it going?” after 30 seconds. This part is often more descriptive, and you won't be able to derive that many new insights. The trick here is to stop yourself from rambling, take a pause, and look for opportunities to go beyond the obvious.
• 2nd step: What interviewers really want are the less obvious insights that require some number crunching. It's almost always the case, especially with tables, that you can multiply/diver/add across to lines/columns to create new meaningful numbers. Align this with your interviewer and go ahead with number crunching - ideally not in silence, so you keep the conversation going. 
• 3rd step: Put the numbers into a context, explain what they mean, relate them to the prompt, client goal, etc. and suggest an idea for how to proceed. Then let the interviewer react and guide.

This is something I train a lot with candidates, and I'd be happy to help you, too!

Best of luck!

Moritz

Hi there,

There is a lot of conflicting advice on a lot of topics - the benefits of having a coach is to reconcile this!

1. If you don't need time to have a good answer, don't take time
2. If you need time to have a good answer, take the time you need

^You don't have a choice…it's a false choice

How do we “hide” not having a good answer? First summarize the exhibit in a sentence. Then, if you need time, ask for a moment. Look for differences in numbers and draw insights from those differences. The moment you're comfortable, speak up!

Hi there,

I would suggest a middle-way.

2 minutes for a graph analysis is too long, 10 seconds likely too short to get the maximum out of it and properly interpret.

I recommend my clients to take 30 seconds to look at the graph to spot they key insights (watch out for hidden insights or combined insights), and their implications.

Taking these 30 seconds also allows you to structure your communication better (top-down, numbered = Pyramid Principle etc).

Cheers,

Florian

I'm not sure what real insight most of us could draw in under 10 seconds on a graph we've never seen before. 

It's always best to take your time and come up with a valuable answer vs rushing and coming up short. 

Definitely approach 1. I would never start discussing a slide with a client before understanding what is there and the key insight.