How To Deal With an Avalanche of Applications (Or, ditch your ATS.)

In 2022 I was laid off from a tech job just before layoffs became the cool thing to do. What followed was a year of job-searching hell that showed me that both traditional and trendy recruiting practices are as well suited to the current tech landscape as a cheap pool floatie is to river rapids. Companies aren't getting what they need, even with a glut of candidates on the market. Candidates can't land an interview, even with companies begging for talent.

It sucks for everyone and it doesn't have to. Join me for a boots-on-the-ground guide to navigating this mess.

Originally this series was going to address things in chronological order, and that meant that this second article would talk about the process of creating a good job posting, determining compensation, and gathering resumes. But then I realized none of that mattered without a fantastic plan on how to deal with a mountain of applications.( And when compensation is intelligently calculated and the job posting is correctly written, there will absolutely be an avalanche of applications.)

In the past, ATS has been implemented, ostensibly to help companies weed out junk resumes and find their best candidates. In reality, ATS sucks. The best description of ATS on the internet goes something like this "ATS can't find a candidate even if they are highly qualified and live next door.” Huck out the ATS. Screen resumes the right way. Get a pair of human eyeballs on every cover letter and resume received. It sounds like absolute madness, in this day and age, with companies reporting hundreds and sometimes even thousands of applications, but don't fear doing it the right way.

A strong game plan will move this mountain very quickly, and make the prospect of creating one so much less daunting. The game plan outlined below will also make application reviews fun. And it may involve free pizza, which is something most people can get behind.

A Basic Mountain-Moving Game Plan

• Assemble the Review Team
• Find a Time and Place
• Print Everything
• Turn the mountain into a pile using lightning-fast early rounds
• Turn the pile into a shortlist using ranking
• Produce a list of top candidates!

Assemble the Review Team: it's best if the team is comprised mostly of the people who will interact with the new employee on the daily, but it's not a bad idea to include a few extras, especially if the mountain of applications for the role is particularly large. For instance, if the team selecting a new Product Manager, grab the developers the new role would be supporting, grab the QA they would be helping, get their supervisor, and maybe yoink a couple of random team members blessed with a good gut from other departments to be your review team. Keep in mind that this is a team where diversity matters, so create as much as possible.

Find a Time and a Place: while this process could be done virtually, it's a process that goes so much better face-to-face. For companies that are usually remote, it's wise to consider this a cross-functional team building exercise that will quite literally build a new team member, and that makes it worth the trip. Regardless of whether this takes place at an office or a meetup, cater good food. Spring for the slightly nicer hotel for those who have to travel. If someone decides that round four is better with beer, desert-disguised-as-coffee, or the spoils of a late-afternoon gas-station raid, they should probably be humored.

Print Everything: there's probably a great way to do this process without a lot of paper, but for most groups it's going to simply work better with paper, so print the resumes and cover letters. Or set up your alternative idea and give it a quick test after reading the steps below.

Round One: A Quick Sort. Give everyone a roughly fair share of the applications. While the end goal of this round is that everyone sorts their stack into “no” and “maybe” piles, the most important goal in this round is getting through as many application as possible as quickly as possible. Encourage people to spend less than a minute or two on each application during this round. Round One will probably reveal a couple of stand-out, phenomenal applications. If these are encountered, put them in their own pile. Reviewing these with the care they are due bogs down the speed, and again, this round is all about leveling the mountain. Also, while it's fine to have a laugh or call out a particularly shiny resume, this round isn't about discussing and collaboratively making decisions.

Round Two: A Quality Check. The team needs to confirm that the “no” pile is, for sure, the “no” pile. Have everyone grab a stack they didn't sort the first time, and go through it and pull out any application they think is a maybe. This second-chance review also helps balance subconscious biases everyone has, and creates a more diverse field of candidates. But still keep it quick and make fast decisions in this round. Again, the time for discussion and collaboration is a bit later.

After round two is finished, everyone should have confidence that the “no” pile is definitely the “no” pile. A good step right now would be to have the no pile candidates get form rejection emails. (It's actually worthwhile to maybe get a temp for the day to help with sending several hundred “No Thanks” emails.)

Depending on how many resumes were in the mountain, and how many reviewers were assigned to the task, this effort will probably conclude around lunchtime. Lunch is a great excuse to go somewhere else, get a break from reviewing application materials, and decompress after the speed rounds.

Round Three: Discuss and Dismiss. By this time the review team has developed some distinct feelings about who/what they do and don't want for the team. Now is the time to facilitate some discussion and collaboration (but don't go too deep, this is still a very fast round!) Collect statements about why people sorted the prior rounds the way they did. One or two reasons per person will suffice. Note them down on a whiteboard for reference and discuss, as a team, if these guidelines work. Right now, the team should mostly be in agreement, or even wishing they'd thought of what their coworker was doing. (If there's a really out there reason for including or excluding candidates, it might be necessary to adjust the approach, but in general, this is just confirming everyone is moving forward with common sense.)

Once the reasons are noted down and easily referenced, do another round of quick decisions on the “maybe” pile from Round Two. Ignore the phenomenal stack for now (though it's fine to add to it.) And yes, confirm the “no” pile produced by round three, just like in round two. This is also when the next set of form rejections can go out.

After round three, the pile should be a pile, not a mountain. Even if the day started with 500 resumes, round three should end with about 50 applications in the “maybe” pile. At this point, some resumes should feel familiar, some reviewers have probably already developed some favorites of this bunch. And that's the perfect setup for the final rounds.

Round Four: Ranking the Semifinalists. In this round, each reviewer should be given more or less the same number of resumes and asked to rank them from best to worst. Encourage the team to discuss with each other when they are struggling to compare candidates. This is the round where it's worthwhile to have the team to share the best and worst stuff they've found. Dramatic delivery isn't optional! Once everyone has a sorted pile, they should then take their top three off the pile and put those top contenders in front of them. Then they pass their less-favorite resumes to the person to their side, right or left. The pile is then ranked again, with again, the top three being held back, and the lower ranks being passed on. Do this rotation a couple of times, but keep in mind that not everyone needs to see every resume. Anyone who hasn't been selected as a top-three contender after a couple of rotations can be sent over to the “no” pile. But hang onto this particular “no” pile. This is your “on hold” pile.

Round Five: Do something else. Go get a great dinner, go do an escape room with the team. Feel free to discuss the day, the resumes, the candidates, what people want in a coworker, what they don't want, red flags, green flags, but don't review resumes or cover letters in this time.

Round Six: The Phenomenals Vrs. The Finalists. If at all possible, do these final rounds on a different day than the proceeding rounds, but not too far apart. The next business day is perfect. The break gives the brain time to digest and maybe have a eureka moment, but is short enough that everything still feels front-of-mind. It's worth starting this day out with a great breakfast. At least go for broke on fresh bagels and a choice of spreads.

In Round Six, t's time to pull in the stack of phenomenals that have been forgotten in the prior rounds. Do a lightning round on them, and then a quality check (essentially a mini round one and two) to see if they still feel like standout, dream candidates. Some of them are probably going to be an obvious no. Put those applications on the “On Hold” pile. The remaining phenomenal applications should be mixed in with the semi-finalists from Round Four.

Round Seven: Candidate Olympics. This final round is the classic group discussion focused on determining which candidates are A listers, and which are the b-listers. On occasion, someone from Round Four will suddenly seem like a much better candidate, which is why the on-hold pile exists. The team should really drill down into what is needed and how each candidate could uniquely fill the job. The aim here is have half or less of the candidates in the a-list.

Once the a-listers have been determined, it's time to start scheduling interviews. But hang onto the b-list and the “on hold” applications until after the first round. Don't send rejections yet. Those two piles are an insurance policy against having to reopen the job and move another mountain. As fun and as team-building as this exercise may be, teams have other things they need to get back to.

Now that the mountain has been proven to be easily leveled, the next thing to address is how to build one in the first place.