Should you switch jobs? - A quantitative decisioning framework

A friend recently reached out to me and sought my advice about a new job offer he had received. Good, reputable firm, decent hike in compensation, same role but elevated title. The role required him to move from Bangalore to Mumbai, but the company was ready to pay for the relocation expenses. The only caveat was that they wanted a confirmation from my friend by the end of the day. No stress, should have been quite a no-brainer to take this job.

Or was it?

While making decisions, especially when we are restricted by time and with too many unknowns thrown in front of us, we instinctively turn to bounded rationality for aid. Bounded Rationality is the idea that our decision making is limited by the information we have, the cognitive limitations of our minds, and the finite amount of time we have.

We also let our cognitive biases influence our decisioning. The new employer can try to anchor our decision by offering a higher title or by offering a higher compensation 'as per industry standards'.

We want to take a rational decision, but we don't know what these levers are and how much they affect the consequences of our decision in first and second order effects.  Given that we are put in the "please confirm by EOD" box, we end up making a back-of-the-envelop decision based on the few available and often unimportant levers. For example -



In reality, there are way too many moving pieces, such as -

• Is the role that of an individual contributor or will you get to lead a team?
• How is the work load in the new role vis-a-vis your current role?
• Is job stability comparable between the two companies?
• How does your career ladder look like in the new company vis-a-vis at your current company?
• What impact will moving cities have on your savings? On your family and kids?
• How is the culture of the new company?
• How realistic will it be to achieve your KPIs in the new company?

... and there could be many more of such variables, or levers, that may impact the consequences of your decision.

These categories and levers can be different for different people. So, it's important (a) to be cognisant of them in the first place, and (b) factor them through quantitative comparison in your decision. So, I set out to make a framework that could help me (or anyone else) take a nuanced decision if they ever find themselves in a similar situation.

The Framework

The basic idea is -
1. Identify the important categories that influence your decision. For me, these were -

• Career Growth - Will you grow faster in the new role or company?
• Company - Is the new company a better fit for you?
• Compensation - Will you be able to meet your financial ambitions?
• Role - Are you a good fit for the role offered?
• Personal variables - Will there be any personal trade-offs for you?

It's also prudent to have a mental picture of how much importance each of these categories should hold in your decision. We'll circle back on why this is important later. Here are the weights that I assigned to each category for myself -



2. Identify and write down the various variables (or levers) that impact the categories you identified in #1 above. Again, these can be different for you and me.

I call these variables "levers" because I intend to use them in this framework as a simultaneously moving system. Think of a gear train in which when one gear moves, all other gears move as well.

This would be a good time to introduce a Google Sheet that I made to put this framework in action. From here, your DIY work is done and all you have to do is gather some data and feed in some numbers. 

3. Once you have identified the categories and levers that are can affect your decision, it's time to assign an importance weight to each of the levers. You can use a simple linear scale of 1-5, where 1 means least important and 5 signifies very important.

Note: If the lever does not matter at all, give it a 0 weight, so that it's removed from calculations altogether. For example, if the new job does not need you to move cities, assign 0 as Lever Importance to the 'Location - City' lever.

The column titled 'Lever Weight' calculates the weighted average % of the importance weights you had assigned to various levers in step 3.

While you work this exercise of assigning weights to various levers, keep looking at the 'Category Weight Mapping' table on the right. This will give you the deviation your assigned lever weights have brought on each category, vis-a-vis the ideal weight you think each category should have.

For example, I thought that 'Personal Variables' should only impact 20% of my decision, but I clearly underestimated this lever. Likewise, I overestimated how much importance I would be giving to Compensation.

[You don't necessarily have to change the lever weights to match your ideal category weights - that would be counterintuitive. The purpose of this table is to bring rationality and awareness in your mental models of decision making.]

4. In the columns 'Current Company' and 'New Company', fill in a linear scale score (between 1-10). Few things to note here -

• Use Data wherever possible (like salary, income, savings, attrition, cost of living, perks etc)
• Avoid Bias. If you think any of your scores are biased, flag those levers and ask someone else about their opinion.
• There could be some levers where you may not get exact data points. In such cases, you could use proxies or surveys to come up with a score. For example, whether the company's culture is conducive may be understood by looking at their attrition rates on review platforms such as LinkedIn or Glassdoor (i.e., a proxy). Or, you could reach out to the last couple of incumbents to understand whether your new boss is supportive, whether the KPIs are reasonable and so on (i.e, a survey). These may not be perfect data points, but they'll likely help.

5. The final thing you have to do is look for the 'Evaluation' matrix and enter the least positive % delta that will convince you to cross the fence. Think of this as your opportunity cost or your risk-reward trade-off.

Once you key in your delta, the framework will compute a weighted score based on the importance levers and your relative scores to each lever for your current company vs the new offer you have. If the +ive delta between the two scores is higher than your threshold, the framework will recommend you to Switch jobs. Otherwise, it will recommend you to reconsider the new offer.

I recommend running at least 3 iterative simulations of this framework to see if you are getting a consistent output. Try adjusting lever weights based on category importance, or refine the input scores for current and new companies by gathering more data. 

Extra framework to include cost of living variables if the new job offer requires moving cities

We often think in first principles. Especially while evaluating a new job offer, the increase in salary is generally set in as an anchor in our minds. In bounded rationality, we confuse this delta with our risk-reward delta (which the framework computes in step #5 above).

Different cities have different cost of living indices. While you might think that a 20% hike is a good number for you to move from, say Hyderabad to Mumbai, you may discount the fact that Mumbai is almost 50% more expensive to live as compared to Hyderabad.

It makes sense to check how much your real income or savings will be if you move cities, and include that as a necessary lever in the framework to make it more effective. I have added a section titled 'Moving Cities' in the Framework to accommodate this lever. You can find the Cost of Living Indices (COLI) benchmarks for different cities (or even countries) on the internet. I used this website as a reference.

Here's a comparison between Hyderabad and Mumbai -

So, even though the fixed salary increases by 20% and the taxes and deductions are less in Mumbai, we find out that effective savings will be lesser in Mumbai than in Hyderabad. These insights may be used as effective inputs in the relevant lever scores.

I hope this framework can help anyone in evaluating a new job offer or in any other decisioning scenarios that involve many changing variables. If you would like give it a shot (or play around with it), please drop me a note at hi@ilipsis.io and I'll share it with you.

Links:
1. Framework: Should you switch jobs? (Also contains a mind-map of various categories and levers)