Consulting Research and Google Senpai

When I first started out in consulting, I was usually tasked with conducting research. In a firm where native English speakers comprised of less than 1% of the total employee population, my command of the global business lingua franca stood out.

Occasionally, this entails conducting interviews costing upwards of US$1,000/hour to gather insights from industry veterans with decades of experience as a first-year analyst. One of my more harrowing experiences include reacting professionally in real-time to a German energy expert as he elaborated on the intricacies of VDE specifications and the policy decisions underpinning them over a phone call. All while my Japanese speaking-only manager looked helplessly at me while stealing quick glances at his watch. The only person who seemed blithely unaware of the time was the energy expert whose enthusiasm for the subject matter and extended compensation could not be told apart.

Most of the times however, research is a much less expensive affair that centers around entering search terms into Google. As much as my peers and I like to joke that we are paid to just Google (and create pretty powerpoint slides), timely and relevant consulting research is a lot more than that. Even though it is treated as a “basic” skill and often relegated to entry-level staff, doing it well requires the mastery of more sophisticated skills.

What is unfortunate is that this skill may not be well-imparted by seniors and managers who are busy fighting their own fires. The result for the poor first-year analyst is pent-up frustration spending needless hours of overtime trawling the web mindlessly for a usable statistical nugget which may or may not exist.

This is what I wish I could have told that first-year analyst.

1. Plan
Before blindly accepting a request to “provide an estimate for the size of XX industry”, or “create a list of top trends that competitors are capitalizing on”, always ask the manager what his/her hypothesis is and what the basis, if any, for coming up with that hypothesis is. Doing so would help you to understand how the research you will be doing can directly support that hypothesis. Not only does that give you a greater sense of purpose to your work, you will also be able to expand your search to include other types of research that can support the hypothesis.

In the course of research, it is always vital to remember who the stakeholders are. Even if the manager explicitly mentions that the research “is not meant to be seen” by the client, the findings (if useful) would eventually be adapted and appear in front of the client in some form. To that end, it is also important to acquire a good grasp of who the client actually is. “Client” is not just an amorphous concept; it refers to actual human beings with their own agenda and proclivities. Your research should always consider not just the direct stakeholder – your manager – but also the end stakeholders – the client. This helps to ensure that the research you do will be relevant to all the people who will be consuming it.

Before entering clicking on that search bar, understand exactly what kind of information you want to collect. If you are planning to collate the research on an Excel spreadsheet, this would mean figuring out what are the fields that you would have and what kind of information you are expecting to populate it with. They range from the factual and simple (“Country : Japan”) to inferential and complex (“Highly likely that Company A is planning such and such due to…”). This preparation will accelerate your work as you can target your search by simply focusing on filling up that spreadsheet based on whatever you can find on the web. Without creating a target beforehand, it is often easy to succumb to link surfing and ride on wave after wave of information that bring you no closer to your goal.

2. Prioritize
Now that your planning is done, it’s time to start relying on Google senpai. To that end, it may be a good idea to prioritize and allocate time to specific search terms. The amount of time your manager gives you to research is highly limited but the amount of content out there is not. Allocate that time to each segment of that research. Chances are, the segments may not be equal. For example, if you are planning to do a deep dive for 3 companies and are given 3 hours to do so, it may be wise to start with Company A (if it is a priority – e.g. has the biggest market share) and allocate it 1.5 hours. The extra time allocated here is to account for learning. After you have created precedence in the form of Company A, it would be much smoother to proceed to Company B and C.

On the world wide web, not all search results are made equal. In your research, you would do well to prioritize the ones that are more important and apply the CRAAP test, designed by Sarah Blakeslee at California State University-Chico in 2004:

• Currency: Timeliness of the info
• Relevance: How the info fits your needs
• Authority: The source of the info
• Accuracy: Reliability and correctness of the info
• Purpose: The reason the info exists

Of the five criteria, Currency and Authority are amongst the easiest to apply at first glance. Specifically, Authority can serve as a good proxy for Accuracy and Purpose. At risk of sounding biased, Gartner is a convenient example in this regard. Known as a leading IT market research advisory, it sources its information from well-connected research analysts and is incentivized to produce high-quality research that it charges per subscription. Quoting Gartner research can take a consultant very far based on my experiences inside and outside of the paywall. But the final and perhaps most important test is still Relevance (touched on later).

Unless you are adept as a speed reader, it is unwise to scrutinize every word in every link that you click. In conducting research, visuals and “Crtl-F” are your best friends. The human brain processes visuals 60,000 times faster than text. The essence of a concept or of the entire article is usually contained within a visual and it is usually far more expeditious to fully understand that visual, be it a graph or schema, than to pore through all the text. In cases where the article is wall after wall of text, it is may be a good idea to do a first cut by reading only excerpts which contain your search term(s). Should an excerpt prove exceptionally interesting, proceed to read the surrounding text to glean for insights.

3. Present
After all that hard work that you have put into planning and prioritizing your research, it would be remiss of you (I would go as far as to call it an injustice), if this was not properly communicated to the stakeholders. Given that research tasks may come up with pressing deadlines, it may not be feasible or expected for you to come up with pretty slides. What is more important is that the research be presented in a structured manner and be easily-understandable. Should you have created a spreadsheet to collate your results beforehand, reusing the spreadsheet as the basis of explanation would often suffice. Other possible means of presentation could be a collection of bullet points on a word processing medium or a rudimentary schema on a sketching software.

Ultimately, the structure and ease of understanding of your work should seamlessly tie back to the original hypothesis and the mandate of the project. Your research should serve as the empirical foundation of whether the hypothesis is true or false, with bonus points for inspiring further discussion among your peers and fueling more hypotheses. In short, the presentation of your research findings should show how it is relevant to the project. It can be as simple as going “The reason why this finding matters is because it proves the hypothesis true and suggests that…[inserts new hypothesis]”. While this may not be expected of a first-year analyst, it is what you need to do well to go beyond being just someone who can type into a search bar and press enter.

While Google senpai traditionally referred to the search engine, in the future it could progressively refer to Google Bard, Google’s conversational GenAI chatbot and a contender to OpenAI’s more popular ChatGPT. Instead of search terms, whole questions can be input with well-formatted results spawned and integrated into productivity tools in seconds. For the same task, the first-year analyst would have to take hours to accomplish. Does this mean that the first-year analyst is no longer relevant?

I would look reassuringly into the frightened eyes of the first-year analyst and tell him that it’s not the case. These new fangled tools may be monopolizing the attention of the business world at the moment. But there will always be a place for the strategy consultant who can get to the heart of wicked problems and navigate the just-as-wicked political maelstroms at a client. And it all starts with knowing how to recognize and conduct good consulting research.