Five Universal Best Practices for Headcount Planning
Podcast Overview
-
Eric Guidice
Headcount Experts, episode five. We're officially past the halfway point. This one's all about the best practices everyone can do when doing headcount planning. Doesn't matter if you're using Meander, the services of Chris Mannion, or Headcount365, the headcount planning software that I've built. These are five things you can do no matter your stage, software, or capabilities as a business. These are the best practices that Chris and I have built across our careers doing it at Wayfair and at Uber. And as a special guest, we have Jim Miller, the VP of People and Talent from Ashby, joining us to give his perspective on headcount planning and how he's doing it today. Very excited for that. Our first guest on the Headcount Experts podcast—what a great way to cross the halfway point.Chris Mannion
Amazing. I feel like we've talked about so many things so far and I'm really excited to learn from Jim about what he's doing, what he's seen in practice, what's changed over the last few months, and how he's thinking about 2026. It's an exciting day.Eric Guidice
Agreed. A wealth of knowledge. We're very happy he's joining, and you'll see him later in the episode. Let's open strong. Here are the five things I think anyone can do in a headcount planning process, regardless of what you're doing. One: make sure your data is clean—whether that's last year's or next year's. We'll get into what that means. Two: have an approval process to change it so you have a record of how changes happen. Three: once you have all your demand in place, map your recruiting capacity against demand. Make a visual representation of that relationship so you can highlight areas where there are gaps or excess capacity. Four: track your variance. If a plan was just the plan, you and I probably would be out of a job. Plans change, businesses evolve, sales happen, people quit. Track all the variance and why it's happening. And five: have an ROI mindset. When you ask for or recommend something, don't frame it as “I.” Frame it in business outcomes: more revenue, more predictable hiring, better execution. Those are my five. Does that resonate? Did I miss anything?Chris Mannion
That's been one of the themes throughout our discussions. I'd drill down on three: data—make sure it's accurate; variance—even with accurate data, if you haven't predicted what's going to happen next year, you'll have surprises; and talking the language the business understands—ROI, cash, revenue, impact. Nail those three and 2026 headcount won't be a breeze, but it will be more successful because you're constantly learning from past iterations. I like the five-point framework. I'd even simplify it further, but it's the right approach.Eric Guidice
People will take what they need from these. The reason I started down the path of headcount was I got good at data cleanliness, which for me meant adding a unique ID—a number that marks the specific line item of demand. It's title-agnostic. A mistake early planners make is tracking by title or department. If people trade, combine, or split requisitions, you need an initial reference point to say, “You wanted a manager of recruiting, but now you have two recruiting coordinators.” How do you tell Finance? Especially at bigger companies with many titles. The unique ID is huge. I've always tracked it as a column so I can see how a requisition or plan changes over time. Have you done unique IDs, Chris? How and why has it been beneficial?Chris Mannion
A unique ID is notoriously hard to track across silos: ATS, HRIS, CRM, engagement tools—each sits in its own data silo. Without a unique ID, it's hard to read across and get useful insights. In the past, we spent a lot of time cleaning data and matching row by row. One simple approach is using something consistent like a personal email address. Those change, but if you want the whole employee lifecycle—from ATS application through employment to exit—you need a unique ID across the chain. Without software to do it at scale, you'll miss data. We often analyzed cohorts separately—ATS funnel, internal employee data, post-exit data—so you tell three stories instead of the full one. It improved over time, but it took a big team and a lot of work.Eric Guidice
A common issue listeners will relate to: the same role posted multiple times. Two locations, two jobs in the ATS, two pipelines, but the same role. Without a unifying ID, you might double hire. A unique ID lets you run the process you want and unify it to a single history that contributes to the future planning process. If you always have one ID, you can track how it changes—plan change rate and all those shifts. The second thing is the approval process: getting a non-invested partner to sign off on a change. Most folks do manager, Finance, executive. There's a lot of complexity. I've seen JIRA, Monday, Google Forms, email, Slack. Use anything—just have a log of someone else approving a change. That protects recruiting from issues like empire building and gives crisp, more accurate plans.Chris Mannion
Often overlooked. Internal tools add process, but do they improve decision velocity and quality? Is the end decision-maker seeing the big picture, or just clicking yes? I've sat with CEOs manually reviewing every requisition because too many were approved and the company was over-hiring. Then everything slows down. It becomes manual and frustrating if you don't have the right system.Eric Guidice
We saw that during the pandemic—over-hiring and rescinded offers. For approvals, in a spreadsheet—or now in Headcount365—I track the reason for change. Why did this change happen? Simple categories: business plan changed, hiring manager changed, recruiting capacity issues. You can get advanced: did approval time hurt the on-time hire? Installing an approval system gives you a secondary audit, a pre-variance checkpoint, so you reduce variance, control manager behavior, track reasons for change, and get a more predictable plan change rate.Chris Mannion
Tie the unique ID into the approval process and you can link backfills vs net-new growth and track talent flow. Then connect it to metrics important for planning: time to hire, cost, quality. How long did a backfill take? If it takes a month to approve, three months to hire, and three months to ramp, you have a seven-month gap. Bake that into next year's plan. Look at what happened this year and join it together with the unique ID.Eric Guidice
And what if you sell a new deal that starts in two to four weeks? That time window is equally important for net new and backfills. Also, good luck tracking merges and splits without a system—combining two roles into one, splitting one role into two, opportunistic hires replacing a role already at offer stage. An approval process helps track those IDs without someone going rogue. Once you have IDs and an approval process, most companies have previous data. Next, map capacity versus demand. What did the business ask last year, and how did recruiting perform? It's tremendously informative and brings executives into the conversation. It supports ROI-based thinking: “Invest in recruiting; get better results.”Chris Mannion
Go deeper: record what you expected capacity to be by team, function, product line, P&L—whatever matters. Compare forecast to actual and the delta. Average those deltas to estimate forecast error, then build it into the next plan. It's operations theory, straight from sales and operations planning. We do this for clients—assume you're as wrong next year as you were historically, then dynamically adjust. Forecast your forecast.Eric Guidice
With Headcount365, we show how capacity-demand data improves performance management—high-performing recruiters or hiring managers asking for more than their process supports. You can start basic—monthly hires and starts—then layer advanced metrics. That data is also subject to variance tracking: variance of production and variance against capacity. My favorite metric is plan change rate—the raw variance of what we thought we'd do vs what was requested in a given year. Year-over-year variance tracking lets you build a buffer and avoid hubris about the plan.Chris Mannion
What surprises did you have last year? Could you have predicted them? If not, expect more surprises this year. It goes beyond data analysis into human psychology. Use what we know about decision-making to improve over time.Eric Guidice
There are good surprises too: promotions that create backfills, big sales deals that require more headcount. Surprise management matters in headcount planning. We have about five minutes before Jim joins, which is a good transition. The last thing: ROI mindset. Leaders like Jim speak in the language of executives. Tie unique IDs, approvals, variance, and capacity vs demand to why it's valuable for the business. How did you do it at Wayfair, Chris? How did you get information in front of decision-makers and get decisions in your favor?Chris Mannion
I inherited our recruiting analytics team and pivoted them to be more operational. We produced tangible data from recruiting that fed forecasting. Running analytics and the recruiting team, I joined the pieces and looked ahead: based on the ask, history, and manager-level performance, what's realistic? Then make recommendations grounded in a comparable team that improved by implementing specific approaches. A product team once posted 70 bespoke roles. Candidates didn't know which to pick. We looked at what worked elsewhere—programmatic hiring and inbound. We consolidated to five evergreen pipelines. I didn't rely on intuition; I showed data on improved time to hire and first-year retention. Compare teams to create psychological momentum and get buy-in.Eric Guidice
We're about to let in the one, the only—Jim Miller. Welcome to Headcount Experts.Jim Miller
Hello.Chris Mannion
Hey Jim.Jim Miller
Thank you. How are you?Eric Guidice
Doing well. Chris and I were just chatting about five things any headcount planning process can do to ensure best practice. We landed on: clean data, an approval process with an external partner, tracking changes and reasons, mapping fixed demand against recruiting capacity, and navigating with an ROI mindset. You're one of the foremost voices in TA today. Do you agree? What's your take? I think some of Ashby's new features on fraud detection and timeline-based inbound give really clean data and enable best practices for headcount planning—regardless of system. For those who don't know you, Jim, please introduce yourself, then let's get into it.Chris Mannion
Yeah.Jim Miller
I'm Jim Miller. I lead the people and talent function at Ashby. We build an ATS among other people and talent technology products. As you can tell by my accent, I live in northeast Georgia in the U.S. Originally I'm from Suffolk in England. I've been in the U.S. for 10 years in January.Eric Guidice
Welcome. It's a pleasure to have you. I'm from New Jersey, which you can tell from the accent. Chris?Chris Mannion
Grew up in the northwest of England, but I've been in Boston a little over 10 years now. My accent is gradually fading, maybe like yours, Jim. Nice to outnumber the Americans today.Jim Miller
Nice.Eric Guidice
All right, Jim—headcount planning best practices. Talk to us about data cleanliness. What do you look for when building your plan for 2026? Share as much or as little as you want. As you're building Ashby's plan for 2026—growth-stage company, strong funding and momentum—what are you doing, when did you start, and how does it come together?Jim Miller
Background: I joined in April 2023, employee 60. We're now 250. Revenue has grown faster than headcount, which is a big positive and a contrast to 2021–early 2022. I don't anchor headcount plans on time. We plan twice a year—September and April—and build a rolling plan anchored to business outcomes, not a time-based model. Outcomes contain time so we can measure quarterly and yearly targets, but we anchor headcount directly to outcomes.For product support, we want fantastic customer experience. Each head is anchored to a total revenue goal. Every time we hit a new revenue tier—say, another two million—one new headcount opens. We're future-proofed. If revenue grows quickly, we don't risk reputational damage by under-hiring behind growth. Conversely, in downturns, roles stop opening because thresholds aren't met. In September we look 18 months out—to end of Q1 2027—and in April, after Q4 and Q1 signal, we refresh another 18-month rolling plan. We're constantly future-proofed against extreme growth or downturns.
Chris Mannion
So you layer milestones and add capacity as you hit them, refreshing every six months. Tactically month to month, do you change those plans or are they locked for that six-month window?Jim Miller
We lock revenue amounts. For startup CSMs, each headcount is based on average book size. If enterprise sales outperform, we may open fewer startup CSM roles. For strategic CSMs, it's based on average book value. We have three headcount buckets: success-driven (anchored to revenue: ARR per head, average book), employee-growth headcount (ratio-based like People and Finance), and investment headcount (AEs, software engineering) where we over-invest to drive everything else. We don't repeat 2021 where the whole company was ratio-based and grew ahead of revenue.We don't “adjust” plans because adjustment is built in. Headcount drip-feeds almost daily. Recruiters open roles continuously. Workload is smoothed across weeks and quarters. Recruiter productivity goes up. Only then do you get clean data on true recruiter capacity. Instead of “eight per recruiter,” we see reality. Our team doesn't break a sweat doing mid-20s per quarter because the workload is spread out. We also do cohort-based hiring with timed postings to influence pipeline volume.
Eric Guidice
For listeners: we've covered rolling forecast vs top-down/bottom-up in our approaches episode. What stands out is spacing requisitions to maximize recruiting capacity. Front-loaded plans are the number one killer of revenue and OPEX forecast accuracy. With predictable demand and strong brand inbound, timed processes make recruiting predictable—the upper echelon of planning execution. How do you factor the time components of revenue? If you're trending toward a target but not there yet, when do you unlock headcount? And how do you plan for recruiter hiring and ramp to meet future demand?Jim Miller
We do offset planning for recruiting—hire recruiters in summer to support Q4 lift into Q1. Align the business on one thing: breakeven. Better quality of hire pulls breakeven earlier. Over-hiring pushes it back. Notice periods, time to fill, and ramp are built into the model, working back in time. It ends up like a just-in-time model where you're always one hire ahead. Evergreen roles: if time to fill is longer than the time to the next revenue threshold, the role becomes evergreen. If we get ahead, we slow the roll.RevOps and Sales Finance predict milestones. If we hit 76, that implies 80 at a specific later point; 76 becomes the trigger to open headcount so the team is in place by 80, with ramp and TTF baked in. Inside Ashby, custom fields capture these dependencies—conditional by function (ARR per head for product support, average book for CSMs). We also mark “sourcing required” and build outreach projects in advance so candidates get contacted the day a job opens. It's data-driven, with interdependencies mapped in the system.
Eric Guidice
Extremely complex. Chris?Chris Mannion
Everything maps to what we've discussed—working backward from the business, predictable demand, dynamic decisions. Many listeners are trying to get to your end-state. What changed over the last two years to get here? How did you move from struggling to close two roles a month to recruiters comfortably handling 20–25?Jim Miller
War stories from 2021–2022. After layoffs in September 2022, I devised the revenue-per model. With FP&A, we replayed history to see whether the layoffs would have been avoided under the model. We found we would have avoided over 100 hires. I had that in my pocket when I joined Ashby. The company already had like-minded philosophies—only add sales when existing reps reach quota, à la HubSpot. Our head of Finance joined soon after—first FP&A hire at Fastly through IPO—so he'd seen it all. We hit a revenue milestone three months early; we compared the planned headcount vs actual and highlighted the delta in customer-facing teams and the reputational risk. That justified a reactive model anchored on breakeven. For anyone starting: anchor on what can go wrong. Your TA leader must be in the room—they've lived this. Put in steps early to future-proof, like Finance pre-approving backfills before departments choose which role to spend it on, so attrition can be your lever to right-size in a downturn. My belief: we should never need a RIF.Eric Guidice
A common thread across us and our customers is partnership with Finance. Companies will do headcount differently. Whether Finance unlocks headcount over time or delivers a bulk annual plan, recruiting leaders should still understand the impact of earlier or later starts and salaries. As you track variance and start dates, you can speak to business impact. How do you communicate performance to the executive team?Jim Miller
We have live, real-time dashboards exposed to the whole leadership team. We've workshopped the plan together, so everyone knows the nuance. Dashboards are built from their perspective—company view or drill into their departments. Deep transparency removes emotion and reduces meetings. We focus discussions on what could go wrong or very right and whether we need to change direction or spin up a new function—like professional services in late 2023. Same page, same data.Eric Guidice
I love it. Data cleanliness, variance tracking, capacity vs demand, and speaking the executive language are the common threads. Chris, closing thoughts or a final question for Jim?Chris Mannion
Any surprises you think people planning for 2026 should be aware of?Jim Miller
AI. I keep hearing about lack of ROI from AI deployed without a plan. There’s organic deployment everywhere. If the tech is designed to save time, that time has to be isolated and repurposed before deployment or Parkinson's Law takes over and ROI goes to zero. In headcount planning, that may mean smaller teams, repurposed work, or repurposed open headcount—do it first, not after. Then assess your risk appetite: deploy the tech or keep humans in seat. One more thing people forget: the cost of lawyers. Savings from AI may be eaten by compliance and legislation costs. Include that in the equation. A customer came to me with a plan; I told them the legal cost. The CEO said, “You carry on—you’re cheaper.”Eric Guidice
A big unknown is what AI will actually do for productivity. In my world, ratios change slightly but not as dramatically as headlines suggest. TBD. Until then, I try to avoid the slop on the apps. Jim, it's been a pleasure. Chris, always great to have the three of us on a topic we've done forever. Here's to many more.Chris Mannion
Thanks, Jim.
Headcount Planning Best Practices Every Company Can Use
Headcount planning is the process of translating business goals into hiring actions, linking Finance, Recruiting, and HR through data and accountability. It defines how plans become execution. When done well, it creates clarity between functions and precision in hiring forecasts.
Best Practice #1 | Data Cleanliness
Maintaining a single source of truth for all position data, connected by a unique identifier across HRIS, ATS, and financial systems, is the critical foundation for accurate future planning. Without clean data, every headcount discussion becomes a debate over which numbers are real.
Unique ID Tracking | Eliminates double-counting, title drift, and reconciliation work. Linking historical plans to current execution data builds the foundation for future forecast accuracy.
Feedback Friday | A unified practice around data integrity where all TA, HR, and Hiring Manager updates to headcount data are conducted to deliver snapshot accuracy periodically throughout the month or the year.
Unified Corporate Taxonomy | Ensure your cost centers, departments, divisions & project naming is consistent across systems.
Utilize a Headcount System | headcount365 gives every headcount a unique ID, and automatically maps all headcounts across the HRIS, ATS, and FP&A system so you have one, always accurate, source of truth.
Jim Miller’s Perspective:
At Ashby, headcount is anchored to business outcomes like revenue and customer growth rather than calendar dates. This allows plans to flex dynamically with performance. Clean, structured data makes that possible by connecting headcount movement directly to business impact.
What happens when you get data cleanliness right?
Greater forecast reliability, reduced reconciliation time, and improved cross-functional reporting.
Best Practice #2 | Approval Controls
Structured governance over all headcount changes through a formal approval process. Approvals act as a “pre-variance checkpoint.” Every role change, delay, or reallocation should have a reason — strategy shift, backfill, or timing adjustment. This discipline prevents plan drift and creates transparency around decision drivers.
Effective approval systems don’t slow the business; they improve decision velocity by reducing ambiguity about who owns each change.
Jim Miller’s Example:
Ashby automates headcount approvals using revenue triggers. As revenue milestones are hit, roles unlock automatically, distributing recruiter workload over time instead of spiking after a large quarterly release.
Impact:
Operational discipline, reduced overhiring, and stronger alignment between business goals and headcount execution.
Best Practice #3 | Capacity vs. Demand
Mapping the relationship between recruiting team capacity and hiring demand to set realistic, data-backed targets. Measuring this relationship prevents burnout, missed targets, and underutilized teams. Historical performance data, such as hiring velocity, conversion rates, and time-to-fill, creates a baseline for how many hires can realistically be achieved. Headcount365’s recruiting leader toolkit enhances functionality in Ashby & other ATS systems by helping TA leaders organize recruiting data “pre-ats”.
Jim Miller’s Perspective:
At Ashby, requisitions are spread evenly across the year instead of being front-loaded by quarter. This steady flow balances recruiter workload, maintains hiring momentum, and provides cleaner data for performance analytics.
Impact:
Increased recruiting efficiency, improved executive alignment, and smoother OPEX distribution.
Best Practice #4 | Variance Tracking
Measuring the difference between what was planned and what actually occurred across start dates, costs, and filled roles. Variance isn’t failure, it’s feedback. Tracking variance exposes systemic issues in forecasting and approvals. Over time, year-over-year variance trends reveal which teams plan effectively and which struggle to execute.
Both positive (growth acceleration) and negative (attrition or delay) variances should inform the next planning cycle.
Jim Miller’s Context:
Ashby uses live dashboards to compare real-time hiring data against rolling forecasts. By identifying divergence early, the team can adjust plans before small misses turn into large gaps.
Impact:
Higher forecast precision, reduced unplanned labor costs, and more adaptive workforce planning.
Best Practice #5 | ROI Mindset
Framing headcount decisions in the language of financial and executive value — not operational need. Finance and executive teams respond to ROI, not headcount volume. Recruiting and HR leaders must translate hiring impact into measurable business outcomes: revenue per employee, productivity lift, or reduced cost of delay.
Benchmarking recruiter output and connecting investments to business goals establishes credibility and accelerates approvals.
Jim Miller’s Example:
Ashby’s people dashboards are designed for executive consumption. They show hiring velocity, cost impact, and productivity outcomes in real time — turning talent data into a continuous business case for investment.
Impact:
Executive trust, faster decision-making, and improved prioritization of resources. As AI introduces new dimensions to headcount ROI, here’s what the experts recommend:
Isolate real-time savings before adopting AI to avoid replacing meaningful work with “busy work.”
Account for compliance and governance costs when modeling ROI.
Adjust future headcount assumptions based on proven efficiency gains, not theoretical automation promises.
Headcount365 Helps Stakeholders Execute a Best-in-Class Planning Process
Instead of spending time organizing data from siloed systems & spreadsheets, headcount365 automatically updates all data in one system. The Request & Approval system tracks all headcount changes, while our AI-powered Headcount Intelligence helps craft this dataset into more accurate future projections.
Finance teams
Integrate rolling forecasts tied to revenue milestones.
Real-time tracking of the plan change rate is a key measure of planning discipline.
Recruiting Leaders
Monitor recruiter workload vs. hiring throughput.
Model realistic capacity and avoid overcommitment.
HR / Workforce Planning
Implement unique requisition IDs and approval tracking systems.
Maintain real-time headcount data hygiene across all systems.
Executives
Review live dashboards that unify headcount and financial data.
Treat headcount as a strategic decision layer, not an administrative metric.
Clean data, transparent governance, balanced capacity, variance awareness, and financial framing turn headcount into a strategic advantage. As automation reshapes how teams plan and hire, these fundamentals remain timeless: discipline, visibility, and alignment between people and performance.