Ask another agency owner what they pay a senior designer and you'll get a number. The number won't help you much.

We've watched agencies find excellent people, in other parts of the US or internationally, at a salary that would be entry-level on either American coast. We've also paid a hefty New York sum for someone whose experience level didn't match the price. There's no settled market rate waiting to be discovered. There's price discovery you have to do yourself.

Before any of this, one caveat worth stating plainly: comp is not what makes talented people stay. Culture matters. Manager quality matters enormously. Whether the work is challenging matters. And the trajectory of the business matters, because people want to be on something that's winning. Nobody's excited about a sinking ship.

But comp is real dollars you're committing, and the more intentional you are up front, the fewer problems you inherit later. Here's how we break it into five layers.

1. Base salary

This is the layer that matters most, for the largest number of your people, and it's the one to nail first.

Start from your own P&L rather than someone else's benchmark. What's your business model, how are you pricing the work, what margin do you actually run? Work backward from there, factoring in utilization and how billable you expect a given role to be, and you'll arrive at what you can afford. Sometimes the honest answer is that you're priced out of a certain tier of talent right now. Better to know that than to discover it in an offer negotiation.

Geography is a real lever here, particularly if you're remote. Finding strong people in different markets for certain roles lets you balance the overall salary load across the company.

Once you've hired a cluster of people, somewhere past 10, 15, 20 employees, you'll naturally start seeing levels: junior, mid, senior, director, maybe VP. That's when salary bands become useful. You grow into bands; you can't start with empty ones and try to fill them.

These don't need to be published to the team. They're primarily an owner's tool, a way of knowing what you expect to pay all the way through.

Pair the bands with a real comp review method. People should know when the conversation happens and what it's based on. Our general shape is a 0 to 5% annual raise. If someone wants a 10 or 20% jump, that's a different conversation, because there's usually a career path and a role to grow into first. Which is also where the individual-contributor-versus-manager question surfaces, and that question changes where the salary can eventually land.

2. Benefits

In the US, health insurance is the big one. It costs the company meaningfully, and candidates weigh it heavily when comparing offers.

The structures vary. Some companies contribute a fixed amount and let employees buy their own plans. Others sponsor a plan and cover a percentage or a set dollar figure. What matters is being deliberate and, if you're a smaller agency still finding your financial footing, conservative.

We all want to offer great benefits. That instinct is right. But costs have trended up year over year, and the more you promise, especially covering families and significant others at a high percentage, the more exposed you are if things tighten. That's a real way to put the business in a tough spot.

There's a self-selection effect too. If robust health coverage is the deciding factor for a candidate, they'll pick the company with the better plan. Sometimes you get the opposite: someone whose spouse covers the whole family through a large employer, which changes the math for that individual entirely.

Retirement contributions belong in this bucket. A 401(k) match, beyond just offering the plan, adds up quickly when you've committed to a meaningful percentage of salary.

So do the ancillary benefits: the milestone gifts, the extra PTO day for a birthday, the anniversary recognition. They cost something, whether in dollars or in time. At Barrel we do a "Barrelversary" gift, an Airbnb gift card at $100 times the number of years someone's been there. By year five or six that's a real vacation, and it's a nice thing to be able to give.

Then there's PTO policy itself, fixed allotment or unlimited. All of it factors into whether someone accepts your offer.

The framing that helps: base salary and benefits together are the comp package. You can run a higher salary with leaner benefits, or a lower salary with richer ones. What matters is which combination is attractive to the people you're trying to hire and sustainable for the business long-term.

3. Bonuses

Two flavors, and they behave very differently.

Periodic bonuses, usually annual, tie to a performance metric. The company hits a milestone, a department hits a KPI. Everyone knows roughly when it's coming and that they'll get something, even if amounts vary. In some industries the annual bonus exceeds base salary, and people build their whole comp expectation around it. That predictability is the feature. It's also the risk, because variability in a bonus people are counting on has an outsized effect on how they perceive their entire year.

Spot bonuses are discretionary and ad hoc. A team over-delivered on a project, the client was thrilled, the company made real money, and you want to acknowledge it in the moment.

Spot bonuses are effective precisely because they don't create the entitlement that periodic bonuses can. And they're delightful to receive; being recognized specifically lands differently than a scheduled payment.

The risk is that discretion can read as arbitrary. Word gets out that spot bonuses exist, and everyone believes they work hard, so where's mine? One way we've seen this handled well is Barrel's employee-of-the-month vote, where the whole company selects someone who went above and beyond. It's still a spot bonus, but it's group-selected rather than handed down, which changes how it feels. Rewards there vary too: gift cards, extra PTO days. There's room to be creative.

4. Variable comp and commissions

This layer is role-specific. It fits most naturally where individual performance ties cleanly to dollars, which usually means sales and business development, where you can point at new business bookings for the quarter or year.

You can get extremely sophisticated here, and it's worth a whole separate conversation. But two things matter more than sophistication.

• Incentivize the right behavior. If you pay a salesperson purely on signing projects, you'll get volume, including projects that aren't a fit and clients who won't stick. Whatever you reward is what you'll receive.

• Show both ends of the spectrum. Set a baseline quota that covers their base salary cost from the company's side, then show OTE, on-target earnings, so the person can see what hitting the milestones actually pays. Minimum and ideal, both visible.

There are more dials: what percentage, for what duration (initial engagement only, the following year, in perpetuity), and whether there's a clawback if a client leaves early.

Extending commissions beyond sales is trickier than it looks. Account managers and project managers can expand accounts, but servicing and retaining the client is already their job. So before you build a comp structure around expansion, ask honestly whether the growth came from intentional effort or from the natural course of doing the work well. When the answer is murky, a spot bonus often fits better. You acknowledge that someone sold a new service line without engineering a formula around what does and doesn't count.

One cash flow note that catches people: tie payouts to when clients actually pay, not when the engagement is booked. That money isn't in the building yet. And avoid backloading everything to year-end. People should feel the momentum as results land, and you shouldn't be staring down a large payout obligation all at once in December.

5. Upside compensation

The long-horizon layer: profit share and equity.

Profit share means setting aside a percentage of profits and distributing it. One approach we've used: set aside a percentage of gross profit, then allocate it by a formula weighted on level and tenure. The longer you've been here, the higher your multiplier; your level carries a factor. Everyone gets a proportionate share of the pool.

Another approach is limiting eligibility to certain levels of leadership, sometimes just the C-suite or the CEO, tying it to the people with the most direct levers on the outcome and making it a significant part of their comp.

Profit share pulls you toward financial transparency, which cuts both ways. If the share is based on profit, people need to understand the profit. We shared quarterly numbers for years: revenue, gross profit, EBITDA, how we were trending. What we learned is that financial literacy varies a lot. For some team members it's just a lot of numbers, and the honest internal question becomes is this good? should I be worried about my job? Open it too broadly without context and you can manufacture anxiety you didn't need.

There's also administrative overhead, and it's heavier than it looks. Running the calculations. Allocating the shares. Writing rules for the awkward cases: what happens if one quarter is profitable and the next is negative, and does the loss come out of the pool proportionally or is the pool protected? Who's eligible at payout, and what about someone who resigned two weeks before? These details only look small until you're administering them.

Equity takes many forms, including RSUs, options, and phantom equity, vested over time or unlocked by milestones.

The theory most owners start from is that equity creates ownership mentality. We'd push back gently. We own equity in Meta and Google; we don't have owner mentality about either. What we care about is the value of the stock. Which means the real question is whether there's a path to that value, most commonly a liquidity path. Is there an exit, and when?

A buyback event is another route. Periodically, let equity holders sell shares back to the company at the current valuation, funded from the balance sheet or by bringing in new investors. Honestly, we haven't seen agencies do this often, though we think it's interesting. There's also the option of letting employees buy in rather than only being granted equity, selling 20% to people who want it, which isn't so different from bringing on an investor except you're offering it to your own team. And there's the ESOP route, a long-term path for transferring the business to the people inside it.

When not to use equity: as a retention tool for a key person you're afraid of losing. We've done this, and we've heard the same story from many owners over the years. You're relying heavily on someone, you fear they'll leave, so you reach for equity as your version of golden handcuffs. Without a clear sense of what it's worth and how value gets unlocked, it's symbolic at best. It might buy you time. It rarely produces the result you wanted.

The right reason is alignment: here's how we get to a successful exit in a defined window, here's how we all pull together, and if we do, you're properly set up to share in it.

Where to actually spend your attention

Three things, in order.

1. Keep it simple at the start, and be realistic about what you can afford. Benefits especially are close to irreversible. Once someone is used to an annual bonus or company-paid health insurance, taking it away is brutal in a way that adding it was not.

2. Get base salary right. For the vast majority of your employees, this is the whole game. A clear comp review system, a real approach to raises, some thinking about bands. That's the 80/20, and that's where most agencies get the most return on the effort.

3. Treat the fancy stuff as optional. Profit share and equity sound great, and they can be great. They also carry overhead you need to be prepared for before you announce anything.