I wanted to share some of my thoughts on what some of our strategies and tactics should be for hiring.

For the first part here, I'll be tackling our apprentice hiring strategy. I think our Mid/Senior Strategy will be quite different and as such, I will address that separately at some point in the future.

# Apprentice Hiring Strategy
I've broken the below into three parts as I think it makes it easier to organize and understand things that way.

The first part is axioms. An axiom is a premise so evident as to be accepted as true without controversy. I don't think there will be much disagreement around the axioms I've described. They just describe reality as it is currently.

The second part is Strategy. You can think of a strategy as a large, overall plan. It orients the direction by which we want to be moving towards. It's possible that there will be disagreement around the strategy. However, it does derive from the axioms, and is just a logical extension of them, so if you agree with the axioms, it's likely you'll agree with the strategy.

The third part is tactics. Tactics are the small steps, tasks and actions we must take to fulfill our strategy. If there is disagreement, it's likely that there will be disagrement around some of the tactics. The good thing is that if you agree with the strategy but not some of the tactics, some of the tactics can be adjusted to still support the strategy.

Alright, here we go:

## Axioms
1. We have a finite amount of resources (time and money) which we can dedicate to recruiting.

2. Successfully hiring a senior developer takes significantly more resources than hiring a junior/apprentice.

3. For every senior developer we hire, we effectively gain capacity to hire one more junior/apprentice.

## Strategy
1. Simplify and streamline our apprentice hiring process. This will free up more resources to focus on hiring more senior developers, which will in turn enable us to hire more apprentices.

## Tactics

### 1. Only focus on a select number of bootcamps.
It’s become very clear that there are certain schools/bootcamps that produce the best graduates. This is evidenced by the fact that all six of our employees that are bootcamp graduates have only come from one of three different schools:

**Flatiron School:** John, Kyle, Edward

**Dev Bootcamp/Metis:** Gustavo, Katherin

**Makers Academy:** Mikhail

Although it’s possible that there are other great graduates from other schools, the amount of time and resources it would take to find them would be significantly more than just focusing on graduates from these target bootcamps.

These target bootcamps should be able to provide a sufficient volume of applicants we will need to make the successful number of hires.

For example,

Flatiron School has three Ruby cohorts each year. Each cohort has 40 students.
Flatiron School also has a Brooklyn campus which puts out around 40 graduates annually.

Total Flatiron grads annually: 160

Every 3 weeks, Dev Bootcamp has approximately 12 students graduate from their program. Over the course of the year, this ends up being around 200 graduates.

Total annual graduates from target NY bootcamps: 360

### 2. Come to a hiring agreement with these target bootcamps.
Agreeing to these bootcamps’ hiring agreement puts us in a better position to hire their best graduates. In particular, we unlock the ability to:
Attend their hiring day and meet the entire graduating class instead of hoping that a student finds us on our own.
Potentially present at these schools, which will improve our brand awareness.

If we don’t come to an agreement with these schools, we leave it up to chance that one of their graduates will find us.

### 3. When we are looking to hire an apprentice, attend one of the target bootcamps’ hiring day.
How we prepare for and attend these hiring days will be critical to our ability to hire the best graduates.

As an example, let’s say we are planning on attending a Flatiron School hiring day. How we approach it should play out something like this:

#### PREPARATION
* Several days before the event, we reach out to Flatiron asking for several copies of their student profile booklets.
* Since Ed, John, and Kyle will be attending the event, they each go through the booklets individually and make note of who they think are the best graduates.

#### AT THE HIRING EVENT

##### First Round (60 minutes)
* Ed, John, and Kyle arrive at the event. Collectively, their goal is to speak with each one of the graduates. To accomplish this, they figure out how they will divy up who they speak to. If there are 40 graduates, then each of them should speak with ~ 13 grads. A simple way to approach this would be to just divide up the room into sections and assign it to each one of them.
* They spend about 5 minutes speaking with each of their assigned graduates. This should take about an hour to do (5 * 13). The goal of each conversations is to get a feel for the candidate, if they would fit with us, as well as get an idea of who the best graduates in their class are.
* During our chat with each of the graduates, the question we always want to ask is, “If you had to choose one person from your class to start a startup with, who would be?” This question is great in that it will allow us to get a very good idea of who are the best people in a given class. In particular, it will tend to reveal the people that best encapsulate our core values as someone who’d you like to start a startup with would tend to have good confidence, teamwork, curiosity, initiative and drive.

##### Regrouping (20 minutes)
* After speaking with all of their assigned graduates, Kyle, John, and Ed get together to regroup and compare notes. Their goal is to determine the people who we should speak with further.
They make note of any people they thought were a good fit.
* They tally up the scores of the startup responses to determine who were the top-scoring graduates.
From this combined list, they decide who are the graduates that we are potentially interested in.
In practice, this will likely yield a list of 30-50% of the initial graduate pool. In this case, it would be 12-20 people.

##### Second Round (60 - 90 minutes)
* Kyle, John, and Ed split up again to chat with the remaining 12-20 graduates that they liked. Each one of them should have the task of speaking with 4 - 7 graduates, and it should not be one of the grads they already spoke with.
* With each of the grads, they should chat for 10 - 15 minutes or so. Again, the idea is to whittle down the list and find out who really would make a good fit for AlphaSights.
* The content of these conversations and the questions we decide to ask will still have to be determined. It’s important to note that these aren’t “interviews”, but are conversations, and our style of questioning and conversation should reflect that.

##### Recap/Review (20 minutes)
* Kyle, Ed, and John meet up once again to go over who they think are the graduates that we should continue to interview. This should likely yield a list of 5 -10 graduates.

### 4. Focus on the best graduates and move them quickly through the interview process.
After determining the best graduates, the goal is to expedite the interview process, so (1) we don’t waste unnecessary time (2) we don’t lose a graduate we would have hired to another company.

In practice, this process will look something like:

1. The day after the event, we personally email the graduates we like. We then begin the process of setting up their interviews. The graduates we liked the most should be the ones we interview first.

2. We invite them for a day of work/pairing. The day will look like:
	* 90 - 120 minutes of pairing in the morning
	* Group Lunch
	* 90 - 120 minutes in the afternoon
	* 60 minutes of interview with Tor

The persons they pair with should change in the morning and afternoon and should not be one of the same people they spoke with at the career fair. At the end of the day after Tor’s interview, the whole team meets up, discusses the candidate and comes to a decision. While this is going on, the candidate can hang out in the kitchen or be on their laptop.

After we decide, we then inform the candidate of our decision that same day. There’s no reason to delay the process. They should know whether we are offering them a job or not when they leave the office.

Note that if the team is unable to come up with a yes by the end of the day, then it’s a No. It’s that simple. Every hire should be an emphatic yes, and any indecision is clearly not an emphatic yes.

### 5. Build relationships with the best graduates we don’t hire.
With this process, there will be some graduates that we liked that we didn’t get to interview or hire. It’s important to build some sort of relationship with them as it is reasonably likely that they will (1) evolve into a strong developer given their initial promise (2) Look for another job at some point in the future. When that eventual time comes when they are looking to move elsewhere, our hope is that they will consider AlphaSights again.

To facilitate this, it’d be recommended that whoever emailed the graduate to set up an interview (which didn’t come to pass because we hired someone else), also follow up and invite them to coffee, offer to provide advice, answer questions, or whatever else. This will leave them with a positive impression of AlphaSights, and make it more likely that they will reapply at some point in the future when they have more experience (and we have more capacity).

### 6. Raise the bar by which we use to evaluate apprentice applicants that come from non-target schools.
The whole point of the above tactics is to hire the best apprentices in the most efficient way possible. Fielding a lot of apprentice applicants from outside of these target schools will effectively negate this benefit of improved efficiency.

Therefore, it makes sense to significantly raise the bar by which we evaluate these other apprentice applicants. As a result, we should have much less first round interviews (and subsequent later round interviews) from this potential pool of applicants.

### 7. Don’t expend resources on trying to hire recent computer science graduates.
This tactic might be a bit controversial as it is in direct opposition to our recent plan and effort to build relationships with certain universities. The crux of it comes down to the following:

**It is significantly more costly to hire a comp science graduate than a bootcamp graduate.**

To understand why, it’s important to ask the question, “How long after a hire’s first day will the total value they contribute be net positive?"

Thinking in these terms, let’s say we hire an apprentice. After their first day, will the value they contribute be greater than their cost?
Of course not. They will require a great deal of onboarding and setup (as would any hire, regardless of experience), and will have contributed very little by the end of the day. (This is totally fine and is not a knock against anyone). If we keep going forward in time, there will be many days where their total cost will be greater than the value they contribute. It will be like this for a while before they get to the point where daily value > daily cost. However, even when this point is reached, the total accumulated value will still be less than their total accumulated cost. As such, it will still take more time before they reach a point where their total value is net positive.

So, with the above in the mind, let’s compare a bootcamp graduate to a computer science graduate across several different areas:

#### Time and cost to hire
**Bootcamp:** Given the more efficient process proposed above, the cost to hire is fairly clear in terms of how to estimate.

For hiring a Flatiron graduate, it would be comprised of the following:

* Initial Student Handbook Screening: 3 hours (3 people @ 1 hour each)
* Going to Science Fair: 12 hours (3 people @ 4 hours each)
* Follow up Emailing: 3 hours (3 people @ 1 hour each)

This provides a baseline of 18 hours to get to the point where we can start doing all-day onsite interviews.

An onsite day will take about 6 hours (2 hours pairing * 2, 1 hour with Tor, 1 hour buffer) of our time.

From there, figuring out the total amount of time to hire a Flatiron graduate, we simply have to calculate how many we expect to have to interview before we hire someone.

Let's say that number is 5. In that scenario, it will take 48 total hours of our time before we hire someone.

One other cost to factor in is that often we will have to pay a ~ $10k hiring fee to the bootcamp.


**Comp Sci:** It's much harder to estimate the cost of hiring a computer science graduate. The majority of it comes down to the large amount of opportunity cost. Here are some things to keep in mind:

* The opportunity cost of the time spent selecting computer science schools that we want to work with.
* The opportunity cost of time spent preparing presentations.
* The opportunity cost of having someone physically travel to go to these schools.
* The actual cost of booking flight and hotel in order to visit a school.
* The opportunity cost of having to construct a process for how we interview and evaluate computer science graduates. It should be noted that the apprentice interview process I described above won't work for computer science graduates. Additionally, comp sci graduates typically struggle with doing our technical challenge up to our standards. This means we will have to come up with a challenge/interviewing process that is tailored to them. This of course will take time and ongoing commitment to improving and iterating upon.

#### On the Job Time to Total Net Postive Value
This one can't be estimated exactly but this is how I would approach it.

The important thing to keep in mind is that a bootcamp graduate will already know Ruby and Rails before they start working with us, whereas a Computer Science Graduate is unlikely to have such knowledge.

**Bootcamp Graduate:**
A bootcamp graduate will will be able to complete smaller tasks on their own. They will still need help from senior developers (which represents an opportunity cost), but it will be less as compared to a computer science graduate.

This is a bit unscientific, but I would estimate that it will take an average of 2-6 weeks before a boot camp graduate's daily value is greater than their daily cost.

After that, moving forward in time, I'd estimate it'd be 4-12 weeks before they reach net total positive value.

**Computer Science Graduate:**
A computer science graduate will require a large amount of time just learning Ruby and Rails. As such, they initially won't be able to contribute much AND will incur a significantly larger opportunity cost from senior developers since they will require more help.

Based upon the above and our experience with Felipe and Eamon (our only relevant data points), I would estimate it would take 10 - 18 weeks before a computer science graduate's daily value is greater than their daily cost.

After that, moving forward in time, I'd estimate it'd be 20-40 weeks before they reach net total positive value. (Keep in mind that they have to recoup their initial negative cost).


### 8. Re-evaluate hiring computer science graduates when our apprentice hiring needs is significantly more than what it is currently.

Continuing the logic from above, I would contend that the effort required to hire computer science graduates is only worthwhile if we are doing so in bulk. The reason being is that we could potentially batch a lot of the onboarding process, thereby reducing the opportunity cost of time spent by senior developers.

So, as an example, if we had 20 more senior developers on our team, and we needed to hire 20 more apprentices/juniors, it could make sense to explore hiring computer science graduates at that point.