Onboarding at Scale

Key Takeaways

• Onboarding friction has direct operational and revenue impact in high-volume hiring environments
• Visibility into candidate progress matters more than automation alone
• Fragmented systems create hidden onboarding bottlenecks and duplicate work
• Candidates who experience onboarding disruptions are significantly less likely to start
• Consolidated workflows reduce handoffs and improve recruiter intervention quality
• TA teams should measure day-one starts, not just accepted offers
• Better onboarding experiences create stronger candidate conversion at scale

Transcript

Note: transcript is auto-generated, there may be typos.

I'm excited to kick off the lightning talks today. But let me first tell you where I come from. So as Jim said, I come from Renuity. I lead our talent acquisition team, and we are in the home services business, which means we're hiring frontline workers all over the United States, usually in remote markets where they don't necessarily have an office that they go in to report to on a daily basis.

And today, I wanna talk about onboarding, and onboarding at scale. Specifically, what happens when you're hiring thousands of frontline workers across those distributed markets. Think field marketers that are going door to door, salesmen that are working in people's homes, and then the installers that are actually doing the work. There's a big gap between accepted offer and day one.

We've gone through a few distinct onboarding eras. What I wanna share today is how we evolved, what broke along the way, and why visibility ended up being more important than automation. Let me start with a real example. Picture Houston. We had 16 offers out, eight people were stuck in the onboarding process, largely because we lacked the visibility on where they were. Especially when it came to things like background checks and pre-employment checks.

The TA team was able to unblock three of those hires, and we had 11 people actually start. Now, that may not sound dramatic, but first, when those hires are tied to training, field coverage, or revenue, there are large downstream impacts. That's really the heart of this talk. Onboarding problems at scale don't just stay in HR. They show up in operations and impact the business For us in Houston, that looked like $400,000 in lost revenue. 

Last year, we had 6,000 accepted offers, but only 5,000 people started. That means 1,000 people accepted an offer and never converted into a hire. And when you hire at scale like that, it's not just noise in the data. That is a huge operational gap. It means the business thinks they have the capacity coming, managers are planning for training, and then a meaningful number of people never make it through the onboarding process. 

I'm gonna walk through in three parts. First, onboarding iterations that we went through. Second, why moving toward one system was a much better visibility. And third, why I believe that talent acquisition has to think about a hire beyond the offer and really own day one starts

To understand where we are now, I'll tell you about where we came from, and for us, it happened in stages. We moved from a fully manual process to a partially automated process, and then finally into a more connected workflow model. Each stage got better, but each stage taught us something different about where onboarding breaks at scale.

In the manual era, onboarding was exactly what it sounds like. For roughly 120 accepted offers each week, that meant 120 emails being sent from TA to an HR representative. That kicked off onboarding. It took more than one hour of manual work for that HR coordinator for each hire. There were about 12 different people involved in the entire process from start to finish. In many cases, this was all happening in our satellite offices who had an individual HR person that was handling all of onboarding. So yeah, it was inefficient, but the bigger problem was that it was inconsistent and very hard to track. There was no central source of truth, just, uh, a lot of handoffs and a lot of individual effort, and not a lot of places for success. 

This slide shows what it actually looked like. A candidate accepted an offer, a recruiter sent an email, we collected candidate information, candidate would reply back, often with sensitive information. We'd then send that sensitive information to an HR person. They would start to enter stuff into ADP. It was all manual. They'd have to build out the job attributes one at a time.

We were chasing the candidate multiple times to complete tasks. That chasing happened before they were in the system and after they're in the system. So the process was not just manual, it was full of stop-and-start moments. Every step depended on a person noticing something, sending something, following up, and checking another system. That's manageable at low volume, but it falls apart very fast at scale.

Our next era was a bit better, but still far from ideal. We were able to reduce the time that it took to enter someone from an hour to 45 minutes by introducing automation and centralizing the process. We set up a web hook that would pull candidate information from Ashby Recruiting into Airtable so we could track onboarding in one place. But the metric on this slide that really stands out the most to me is candidates who were rescheduled because they had incomplete onboarding never actually started. So that tells you onboarding friction is not just an administrative inconvenience. Once a candidate hits a disruption in the process, a large share of them are gone for good

In this era, when we move forward, we're still patching it together. As I said before, the webhook that pulled information out of Ashby and into Airtable, the payload didn't include all of the data that we needed. We had to go back, find the missing information in Ashby, and enter it into Airtable so we could eventually then reenter everything into ADP

Some of the communication was automated coming out of Airtable, but it still relied on someone clicking the right button at the right time. So this era was better than the fully manual era, but it was still fragmented. We had fewer spreadsheets, we had more consistency, but we still had handoffs and duplicate work, limited visibility into where someone really was between their offer and their start date

And this is a shift that changed the conversation for us. At a certain point, we realized the issue wasn't just that there was too much manual work. The issue was operating across too many systems and still couldn't clearly see where the candidate was, and they were getting stuck. So it became bigger than automation.The problem was visibility.

In our current era, we move toward keeping the experience in one connected flow, from Ashby Recruiting to Ashby People Workflows, and then into ADP. We partnered with Ashby as a design partner on both the onboarding workflows and the ADP integration, so candidates can move directly through the process instead of bouncing between multiple systems.

That matters operationally, but it also matters for candidate experience. The candidate is already familiar with the Ashby platform. It's where they applied. The interface is much easier to use, forcing them into, rather than forcing them into a more laborious ADP UI. For our hiring population, which as I stated is largely blue collar, frontline workers in distributed markets where we don't have someone on site who can hold their hand and help them, reducing that friction is critical. The harder you make it for someone, the less likely it is that they're gonna actually start 

Operationally, the numbers here tell a story. We went from hundred, over 100 emails a week to kick off the onboarding process to zero. We went from 12 team members involved to four. That means less coordination, fewer handoffs, and fewer chances for work to stall between teammates. And just as important, it means as a team, we can spend less time moving information around and more time helping candidates through the process.

This is one of the clearest examples of what visibility actually means. Before, background check status was scattered. We were checking one system in another, and it's still not always getting a clean answer. Now, with better workflow visibility, the team can see much more clearly where someone is stuck in the process and specifically what's blocking them. That changes the quality of intervention. Instead of chasing someone, calling them, and saying, "What's wrong?" We now can say, "I see exactly where you are, I know exactly what you're stuck on, and I know how to help you." That's better for the team, but more importantly, it's better for the candidate.

Owning the hire is a mindset shift that I care about most. Too often, talent acquisition is measuring hires and finishing the job when an offer is accepted, but the business doesn't get any value from accepted offer. The business gets value from starts. They get value from people actually showing up, and coming to, coming to training and getting productive. For me, it's about owning the hire, not just owning the offer. 

And that leads to my final point. More automation gives less context switching and more direct candidate engagement. The win's not just efficiency. The win is that we can spend less time chasing candidates and more time creating a better human experience. And especially in frontline hiring, it matters a lot. If your onboarding is hard, confusing, impersonal, candidates are just gonna go find a smoother, easier job. 

So if there's one thing that I leave you with, it's this. If you're onboarding at scale, don't ask whether the process is automated, ask if it's visible, if it's simple for candidates, ask whether your team can intervene before someone drops, and ask whether measuring accepted offers or you're measuring actual starts. 'Cause a job isn't done at offer accepted. The job is done when that person shows up and is set up for success. Thank you.