As we have explained in a previous blogpost, service-based businesses are typically characterized by a pendular cycle, and the processes of hiring and selling sit at opposite ends of this cycle: the more projects we sell, the more people we need on the team, and the more people we have on the team, the more projects we need to sell. This isn't necessarily the way to go about business, but most organizations do it following this pendular pattern, going from one to the other as each gains urgency.
This pendular motion can quickly become problematic, and in our attempt to circumvent the problems associated with this, we've strived to create a virtuous cycle , which has enabled our company to survive and thrive for almost 20 years.
At the beginning of 2022, we entered a new cycle of growth, as we decided that we wanted to create more slack for our internal processes. This meant that we needed to hire more engineers in order to generate more capacity to hire other non-billable roles that create wellbeing for our teams. In a dynamic and saturated industry such as software development, this presented a challenge. And we love a good challenge.
So, we began by identifying some of the problems to be tackled :
- We needed to fill 9 roles (at least!) with not a lot of urgency, but most of them with high priority:
- Engineering roles (but pretty much every role associated with tech) are in high demand worldwide, and there is not an abundance of available people with the level of skills we require.
- Due to #2, there has been a proliferation of recruiting efforts across the industry, and the competition is fierce.
- Due to #2 and #3, recruiters have become more and more aggressive in their approach, and that has led engineers to raise their inbox entry barriers. Most of them don't respond, some of them aren't even findable.
- Since we normally hire very organically and at a low pace, no one at Manas is fully dedicated to this task, so it's carved out from other roles' time.
Associated with these dimensions of the problem, we can identify a series of assumptions. That is, the biased perspective with which we approach thinking about these problems. Most people would agree that biases are a bad thing, but instead of contemplating how those affect their perceptions and decision-making, they ignore biases altogether and end up becoming even more biased. These were the assumptions we identified at the start:
- There is too much noise : as we described earlier, too many organizations are looking to hire technical roles, and there are not enough engineers to go around. So, if we're going to reach the right people, we need to find ways to stick out. This, we believe, will likely be easier if we move outside of the usual channels (email, LinkedIn, job boards).
- Outbound only works when expected : Direct outbound messages only work when there is an actual expression of interest from a candidate in finding a new job (maybe they've posted about it on social media, or we've learned about it from a direct connection).
- When that interest in a new job is public, there is blood on the water.
- Culture is hard to communicate externally : From the outside, it's hard to see what goes on at Manas at the culture level, which is one of the things that makes people join us, and then stay at Manas for so long.
- Being an outlier is difficult to explain : Most job titles sound boring, watered-down and conjure images of bureaucracy, which are far from what really goes on at Manas. However, the actual lack of a managerial hierarchy at Manas (and consequently not a lot of emphasis on job titles) can come across to some as too unstructured, even sketchy.
- The devs we want are not in plain sight : Most developers don't use LinkedIn (or any of the popular job boards, for that matter), or they do but have gotten used to leaving tens of messages unread. In practical terms, we can assume they don't use it (not in a way that allows us to reach them). However, they can be found on Twitter, Reddit and other, more obscure forums, where it is not readily disclosed what they do for a living, their experience, who they work for, etc.
The junction of problems and assumptions identified enables us to formulate a few hypotheses to frame our approach:
- Among senior, high qualified engineers, there is a segment that is no longer attracted merely by high compensation, but is looking for something else:
- Company mission
- Interesting tech stack
- Autonomy
- Stimulating projects
- Culture
- There is a chance that our culture can be effective in luring senior developers, so we need to find ways to make that visible.
- The best shot we have to reach engineers is through their peers (social proof).
Actions
Having mapped what we know about the problem and the assumptions that we are operating with, and having formulated our hypotheses about potential solutions and opportunities, we can begin outlining a series of actions that can address these problems:
-
We developed a landing page with the roles we were looking to fill, where we could also showcase some of the benefits inherent to our culture and way of working. This also gave us more insight and control over the pipeline, without having to rely on external tools like social media or job boards.
-
Then, we encouraged everyone at Manas to share our openings with their network, in their own terms and with their own voice, to test our third hypothesis
- We also tried to get people in the Manas network (clients past & present, partners, friends, alumni) to tweet/retweet about our openings, which also shone a light on the fact that we're in the midst of a healthy growth spur
-
On the outbound side of things, we opened up two fronts:
- We reached out to a few profiles that we were interested in, trying to be as specific in our approach as we could, crafting proposals that would be relevant to each person
- For those who politely declined, we politely asked if they'd be up for sharing our search with their network, or recommend us similar profiles
- After conducting some research, we hired a talent agency that helped us increase our reach and maintain a higher interview pace
-
Trying to avoid the clutter in the usual job search channels, we researched and joined groups, boards and communities where we hypothesized engineers might gather
- There, we were very careful to promote our openings respectfully and in the designated channels
- Joining conversations, we identified opportunities to speak about some of our highlights as an unconventional organization (our handbook and other internal tools that usually are of great appeal to potential candidates) as long as they brought value to the conversation, and never as a way of merely talking about us.
How it played out
Over the course of three months, we led nearly 90 interviews, out of which 16 ended in offers and 8 in effective hires. It may sound like a small number, but bear in mind that we are a relatively small company, and those 8 hires translate to a 30% increase in our staff.
One of the interesting new facets of our hiring process is that over the past years, we went global, opening offices in three cities besides our hometown of Buenos Aires (Montevideo, Seattle and Toronto), and hired people in several other countries. This translated to an increase in our options for available candidates, but also meant that we had to streamline our operation on the administrative end.
Another challenge that drove us to rethink our internal organization came up around technical interviews: as these require technical staff, it means carving a significant amount of time out of our engineers' schedule, which proved difficult. Of course, this was one of the original reasons why we wanted to hire more people, to create some wiggle room in everyone's day. But the transition proved tricky in terms of coordinating multiple availabilities, sometimes across 3 or more timezones.
There is a substantial workload that goes into sourcing profiles that match what our organization needs at the moment, recruiting them into the interview process, which in turn requires a hefty amount of management –and some luck– and then, if all went well, moving candidates up into the onboarding pipeline.
The keys to this entire process were agility, transparency and collaboration. It takes a lot of coordination effort to structure an efficient, multithreaded recruiting pipeline. Without it, it would've been impossible to hire as many people in such a short timespan, not having a full time dedicated team. And there is still work to be done.
Check out our current openings, and if none matches your profile, maybe consider designing your own role. In the end, that's what Manas is all about.