Playbook
The Post-Raise 90-Day Back Office Playbook
What to set up, in what order, with which tools. Built for seed and Series A founders who closed last week and want the back office handled before the next board meeting. Specific tools, specific prices, specific traps.
Jump to the checklistThe window
90 days is the only good window
The 90 days after a raise is the cheapest the back office will ever be. You have momentum, capital, and clarity about what is broken. Three months later you will be pattern-matching to whatever ad-hoc fixes you put in, you will have hired around the gaps, and the cost of pulling those threads will have tripled.
Most founders use the post-raise window for hiring product and GTM, which is correct. The mistake is leaving the operational layer to whatever comes next. Whatever comes next is usually a Head of Operations who arrives in month four with a headcount plan, a vendor procurement habit, and no system to inherit. That hire ends up building from scratch on a foundation no one designed, which is the most expensive way to get a back office built.
Build the foundation in the window. Then anyone you hire walks into something that already runs.
This playbook is what we install during a Headroom engagement. Same order, same tool selections, same opinions. If you would rather run it yourself, the next 3,500 words are exactly that. The interactive checklist further down is the same one we work from, with progress saved locally in your browser so you can come back to it across the 90-day window.
The 90 days after a raise is the cheapest the back office will ever be.
The stack
One stack, configured the same way every time
There are nine functional surfaces post-raise: banking, corporate cards, payroll, contractors, accounting, equity, sales tax, HRIS, and compliance filings. Most founders try to evaluate each one in isolation, get pitched by five vendors per category, and end up with a stack stitched together over six months by whoever was available.
Skip that. The decision is mostly already made. Below is the stack we run for every Headroom client at this stage, with the specific reasons each tool wins. Where there is real optionality (Gusto vs. Rippling, Pulley vs. an existing Carta footprint), the picking rule is in line.
The principle: every tool needs to be modern, API-first, and willing to talk to the others without a paid integration tier. Anything that fails one of those three is out, regardless of price.
The tools
Tool by tool: what wins, what falls short, who it is wrong for
The summary above is the recommendation. The sections below are the reasoning. Each tool gets a dedicated treatment with strengths, weaknesses, pricing at scale, edge cases, and the company profile it does not fit. If you only read one section, read the one that covers the tool you are about to commit to.
Pricing is current as of May 2026 and changes. Treat the numbers as directionally correct rather than quote-grade.
Banking
Mercury
The pick: Mercury for the operating account, full stop, for any company under $50M in operating cash. The decision is over before it starts.
What Mercury does well
The interface. Mercury is the first business banking experience in two decades that does not feel actively hostile. Wires take 30 seconds to initiate. ACH is one click. Sub-accounts behave like sub-accounts, not nested entities with separate logins.
The API. Mature enough to power custom workflows without a third-party middleware layer. We use it for automated payroll funding sweeps, vendor payment automation, and real-time cash dashboards. None of the legacy banks offer anything comparable.
Treasury yield. Roughly 4.3% APY on idle cash through Mercury Treasury, swept into a money market fund daily. At $5M of operating cash, that is about $215,000 a year in yield you would not get at Chase or BofA. Cost to enable: zero.
FDIC sweep. Up to $5M of FDIC coverage through their partner bank network. Meaningful in a post-SVB world where deposit concentration risk is a real board-level conversation.
Integrations. Native sync with QBO, Ramp, Brex, Stripe, Rippling, Gusto. The ones that matter all just work, and the data flows are clean enough that reconciliation does not become its own subproject.
Where Mercury falls short
Cash deposits. If your business takes physical cash (rare for our client profile but real for restaurants, retail, and physical event businesses), Mercury is wrong. There is no branch network and no cash deposit infrastructure.
Credit and lending. Mercury Capital exists, the offerings are limited, and the underwriting is less friendly than a relationship bank would be at scale. If you expect to need a $5M+ revolving credit facility within the next 18 months, build a parallel relationship at SVB or HSBC.
Lockbox for paper checks. If you have customers who insist on mailing physical checks (common for enterprise sales, government contracts, some medtech), Mercury's lockbox is third-party and clunky. JPMorgan Chase or BofA Treasury Services are better here, even with the UX cost.
Foreign currency holding. Mercury supports outbound foreign wires but does not hold non-USD balances in any meaningful way. If you have a UK or EU entity collecting GBP or EUR, you need a Wise Business or Revolut Business account in parallel.
Pricing at scale
Mercury is one of the only stack components that does not scale in cost. Treasury yield offsets every other tool you pay for in this playbook combined.
Edge cases
Brex Cash as the alternative. If you are committed to the Brex card stack and want a single banking surface, Brex Cash is fine. Pays interest, syncs cleanly with Brex spend, does the basics. We do not recommend it as the primary because the UX is worse, the API is less mature, and Brex's banking partner has had recent volatility. Not wrong if you are already there. Do not migrate just for our preference.
Keeping a relationship bank. If you have an active SVB or First Republic line of credit, do not break that to chase Mercury's UI. Run Mercury for operations, keep the relationship bank for credit. Many of our clients do exactly this.
Who Mercury is wrong for
Cash-heavy retail. Lockbox-dependent enterprise sales with paper-check-only customers. Companies with strong existing SVB or First Republic relationships and active credit lines. Companies whose primary banking need is foreign currency holding.
Headroom verdict
Default. The decision is made. Spend your evaluation cycles on tools where the answer is actually contested.
Corporate cards and spend
Ramp
The pick: Ramp for corporate cards and spend management for any company at this stage. The strongest single recommendation in this entire playbook.
What Ramp does well
The policy engine. Ramp's spend policy lives at the card level, enforced at swipe. Set a $250 monthly software cap on an employee's card and the 11th transaction over the limit declines automatically. No after-the-fact flagging, no expense report dispute, no founder review. Policy is the system.
Vendor consolidation. Ramp surfaces every SaaS vendor the company pays for, by user, with last-login and usage data pulled from connected platforms. We have not run a vendor audit through Ramp where the company did not find $20,000 to $80,000 of annualized waste in the first month. The audit pays for the rest of the stack.
Bill Pay. Ramp Bill Pay is the AP module: invoice ingestion, approval workflows, ACH or check disbursement, accounting sync. Replaces Bill.com cleanly. Pricing is per-payment with a generous free tier; Bill.com bills per user per month and starts at $79 per user.
QBO sync depth. Ramp pushes line-item detail to QuickBooks with your COA mapping, including categorization, tax treatment, and project codes. Most card platforms push a single GL line per transaction. Ramp's sync is the cleanest in the category.
No upsell pressure. Ramp's sales motion is consultative. Their support actually solves problems instead of routing you to an account exec. The contrast with Brex (mid-pivot toward enterprise) and Bill.com (legacy enterprise sales motion) is significant once you experience both.
Where Ramp falls short
Travel booking. Ramp Travel exists. The inventory and the UX are weaker than Navan or direct booking. If your company does meaningful business travel, run it through Navan or direct and let Ramp handle the card.
International card issuance. Ramp issues virtual cards in USD only. If you have an international entity that needs to issue cards in local currency to local employees, Ramp is wrong for that entity. Pleo or Soldo cover EU. Brex has slightly better international coverage at the high end.
Out-of-pocket reimbursements. Ramp handles employee reimbursements but the workflow is less polished than Expensify or Rippling's reimbursement module. Workable, not best-in-class. Most clients use it anyway because the consolidation benefit beats the UX gap.
Pricing at scale
The free-for-life model on the core product is the single biggest wedge against Brex. Brex started free and has been quietly tightening into paid tiers for years. Ramp has held the line.
Edge cases
Brex as the alternative. If you have a Brex Cash account already and want one vendor for cash and cards, Brex is fine. The card itself is competitive. The reason most clients land on Ramp despite that is the trajectory: Ramp is investing in product depth at the post-raise stage; Brex is climbing toward enterprise. Two years from now Ramp will have more depth for your size; Brex will be further upmarket.
Mercury IO as a third option. Mercury launched its own card product. Acceptable for very small teams that want one vendor. The policy engine is significantly thinner than Ramp's. Not the right answer past 10 employees in our experience.
Who Ramp is wrong for
Companies whose primary spend is international card issuance to local employees in non-USD. Companies whose AP volume is dominated by international wires (use Wise Business in parallel regardless). Cash-heavy operations with daily cash deposit needs. Companies where the founder has a strong personal preference for a Brex relationship (the gap is not large enough to fight over).
Headroom verdict
Strongest recommendation in this playbook. Issue Ramp cards on day one. Skip Brex unless you have a specific committed reason to be on their stack.
Payroll (US-only, simple)
Gusto
The pick: Gusto if you have under 50 W-2 employees in three or fewer states, no global team, no equity comp complexity, and no strong HRIS or device-management needs.
What Gusto does well
Setup speed. End-to-end onboarding takes 48 hours. The wizard is good. We have set up Gusto for 15-person companies in a single afternoon, including benefits enrollment.
Multi-state US W-2. Gusto handles state registrations themselves. They also automate the ongoing reciprocity, withholding, SUI, and unemployment filings without intervention once the registration is complete.
Benefits. Gusto's benefits broker arm is competent and the medical, dental, and vision plan options are reasonable for small companies. Not best-in-class, but inside the platform with no separate broker relationship needed for the first 50 employees.
401(k) integration. Gusto's Guideline integration is one click. Setup, contribution, compliance testing all handled. This is real friction reduction for a benefit that most founders otherwise put off for two quarters.
Pricing. Plus tier at $80/month + $12 per employee per month. For a 25-person company that is $380/month all-in. Rippling at the same headcount runs $1,500+. The cost gap is real and matters when you are 18 months from the next raise.
Where Gusto falls short
International. Gusto does not run international payroll at all. You will need Deel or Rippling Global on top once you hire abroad. This is the single most common reason clients migrate off Gusto.
HRIS depth. Gusto's HRIS module exists and is acceptable through about 50 employees. Past 50 you will want either Rippling, BambooHR (overpriced for what it does), or HiBob (better at scale).
IT and device management. Not a feature. If you want zero-touch laptop provisioning tied to onboarding, Rippling is the only option in this category.
Equity comp depth. Gusto handles RSU and ISO payroll mechanics but the integration with cap table tools (Pulley, Carta) is weaker than Rippling's. For complex equity events at scale (acceleration, secondary, large-grant payroll tax), this matters.
International contractors. Gusto pays US-based 1099 contractors only. International contractors need Deel or direct-payment workflows with manual 1099 tracking, which is friction you should avoid.
Pricing at scale
Edge cases
When to start on Rippling instead. If your day-one hiring plan includes employees in 5+ states, an international entity, or 1099-to-W-2 conversions for offshore contractors, start on Rippling. The Gusto-to-Rippling migration cost at year two is real (4 to 6 weeks of internal time, around $15K of engineering and ops cost). Better to commit to the heavier platform if you will need it within a year.
Justworks as a PEO alternative. Justworks takes legal employer status (PEO model). Acceptable for very small companies under 25 where the founder wants to outsource HR liability entirely. Tradeoff: lose flexibility on benefits selection, take on PEO migration complexity if you ever leave. We rarely recommend it because the lock-in is severe and the cost-benefit reverses past 30 employees.
Who Gusto is wrong for
Any company that will be international within 12 months. Any company over 50 employees. Any company with serious equity comp complexity. Any company that needs zero-touch device provisioning. Any company hiring offshore engineering on a contractor model.
Headroom verdict
Best-in-class for the simple US-only post-raise company. The default for seed and Series A unless something in your hiring plan rules it out within 12 months.
Payroll, HRIS, and IT (complex)
Rippling
The pick: Rippling if you have employees in 5+ states, plan to hire internationally within 12 months, want HRIS plus payroll plus IT in one platform, or need depth on equity comp payroll.
What Rippling does well
One source of truth. Headcount, payroll, benefits, devices, app provisioning, time tracking, and expenses all in one platform with a single record per employee. The data model is materially better than the cobbled-together alternatives. When someone is hired or terminated, every downstream system updates from one trigger.
Global payroll. Rippling Global runs payroll in 50+ countries with local entity setup. If you have a UK or Canadian sub, Rippling is the cleanest single-platform answer. Tax filings, benefits, and local compliance all handled inside the platform.
Device management. Zero-touch laptop provisioning (MDM), automated app onboarding via SSO and SCIM provisioning, device retrieval workflows on offboarding. Replaces JumpCloud or Kandji at small scale.
Workflow automation. The Workflow Studio allows custom triggers ("when an employee is hired in California, do X, Y, Z"). Useful for bespoke onboarding sequences that would be manual elsewhere.
Equity comp depth. Rippling handles ISO, NSO, RSU, and ESPP payroll mechanics with cleaner cap table integration than Gusto, particularly on the Pulley and Carta sync. Complex events (option exercises, secondary, large RSU vests) work without manual journal entries.
Where Rippling falls short
Setup time. Implementation takes 2 to 4 weeks vs. Gusto's 48 hours. The platform is broader, the configuration is heavier, and the learning curve on the admin side is real. Plan accordingly.
Pricing complexity. Rippling sells modules. Base payroll is one price, HRIS is another, IT cloud is another, spend management is another. The all-in cost surprises companies that priced only the payroll module on the sales call.
Customer service. As Rippling has grown, customer service quality has degraded. Email response times of 48 to 72 hours are common. For a 15-person company on Gusto this would be unheard of.
Lock-in. The unified data model that is Rippling's biggest strength is also a switching cost. Migrating off Rippling is materially harder than migrating off Gusto. Plan for this when you commit.
Aggressive sales motion. Rippling's account team will find every module they can sell you. Push back on anything you do not need. The default experience is to be sold up; the right experience is to buy what you need.
Pricing at scale
The price gap with Gusto is roughly 4 to 5x. The depth gap is roughly 4 to 5x as well, but it only matters past 50 employees or with international operations.
Edge cases
Skip Rippling for Justworks. Justworks is a PEO; they take legal employer status. Legitimate alternative to Rippling for very small companies under 25 where the founder wants to outsource HR liability entirely. Lock-in is severe past 30 employees, so we rarely recommend it for our client profile.
Run Gusto plus Deel as a parallel stack. If you want the simplicity of Gusto for US payroll and Deel for the international side, that combination works. The downside is two platforms, two reconciliations, two onboarding flows. Acceptable for under 5 international employees, breaks past 10.
Who Rippling is wrong for
Simple US-only companies under 50 employees with no international plans (use Gusto). Companies that prioritize fast setup and minimal admin overhead. Companies whose founders hate dealing with a multi-touch sales motion. Companies that already migrated off Rippling once and have institutional pain attached to it.
Headroom verdict
The right answer once any of the trigger conditions are met. The wrong answer to default to without those triggers, because the cost and complexity overhead is real.
International contractors and EOR
Deel
The pick: Deel for international contractor payments, employer of record (EOR) services, and global hiring without setting up a local entity.
What Deel does well
EOR coverage. 150+ countries. If you want to hire one engineer in Argentina or one designer in Portugal without setting up a local entity, Deel hires them on your behalf. Payroll, taxes, benefits, and compliance all handled inside their local entity.
Contractor payments. Deel pays international contractors in their local currency, generates compliant invoices, and handles 1099 and W-8 paperwork at year-end. Faster and cheaper than running international wires through Mercury for the same volume.
IP assignment and jurisdiction-aware contracts. Deel's contractor templates include enforceable IP assignment clauses by jurisdiction. Most generic contractor agreements have IP clauses that are unenforceable in countries like Germany, Brazil, and Russia. Deel solves this at the template level, which prevents an expensive legal problem two years later.
Equity for international hires. Deel can administer option grants for employees in jurisdictions where direct US equity is tax-hostile (UK, France, Germany). Handles EMI/qualifying-option treatment where applicable. The integration with Pulley and Carta is reasonable.
Where Deel falls short
US W-2 payroll. Deel does not run domestic US W-2 payroll well. Use Gusto or Rippling for the US side and run Deel for international in parallel.
EOR pricing at scale. EOR runs $599 per employee per month. Once you have 5 to 10 employees in a single country, setting up your own entity becomes more cost-effective. Deel's own consulting arm will help you transition off their EOR onto your own entity, which is unusual and to their credit.
Reporting and accounting integration. Functional but not deep. You will spend more time reconciling Deel-to-QBO than Rippling-to-QBO. Build the close calendar with this overhead in mind.
Customer service. Has gotten worse as the company has scaled. Response times of 24 to 48 hours on contractor-side issues are common.
Pricing at scale
Edge cases
Deel vs. Remote.com vs. Velocity Global. Deel is the broadest in country coverage. Remote.com is the cleanest UX. Velocity Global is the most enterprise-grade for 50+ international employees. For our client profile (under 60 total employees with a few internationals), Deel is the default. Reconsider at 50+ international.
Rippling EOR as the alternative. If you are already on Rippling for US payroll, Rippling EOR is a legitimate alternative to Deel for international. Coverage is narrower (about 50 countries vs. Deel's 150+) but the integration with the rest of your stack is tighter.
Who Deel is wrong for
Any company without international contractors or employees. Any company using EOR for 15+ employees in one country (transition to your own entity). Any company with strict data-residency requirements that Deel's structure cannot meet.
Headroom verdict
Default for international contractors and EOR. The category is contested but Deel is currently the most mature for our client profile.
Accounting
QuickBooks Online
The pick: QuickBooks Online (QBO) at the Plus or Advanced tier. The boring, correct answer for 95 percent of post-raise companies through Series B.
What QBO does well
Accountant ecosystem. Every bookkeeper, fractional controller, and tax accountant in the US speaks QBO. Hiring an accountant is materially faster on QBO than any alternative. The hire-and-onboard time gap is two to four weeks compared to NetSuite or Sage Intacct.
Integration breadth. Native or near-native sync with Mercury, Ramp, Brex, Stripe, Gusto, Rippling, Deel, Pulley, and Carta. The integration surface is the broadest in the category. Whatever stack you commit to, QBO talks to it.
Pricing. $90 to $200 per month all-in for the platform. Compared to NetSuite ($50,000+ per year first-year fees), Sage Intacct ($25,000+ per year), or even modern entrants like Rillet ($1,000-2,500 per month), QBO is the cheapest functional answer at this stage.
Mobile and AP. Mobile receipt capture, vendor payment workflows, and bank reconciliation work well enough that a non-accountant founder can operate the platform for the first six to twelve months without a dedicated bookkeeper.
Payroll integration. QBO Payroll exists and is fine for very small companies. For our client profile we recommend running Gusto or Rippling separately and syncing the journal entries in. Cleaner separation of concerns and stronger payroll feature set.
Where QBO falls short
Multi-entity consolidation. QBO does not handle multi-entity consolidation natively. If you have a US entity and a UK sub, you will run two separate QBO files and consolidate manually (or in a tool like Joiin or Numerik). This is the number-one reason companies eventually migrate to NetSuite.
Inventory. Basic inventory only. If you sell physical product with SKU complexity, you will need a separate inventory system (Cin7, Katana, Fishbowl) syncing to QBO.
Audit trail depth. QBO's audit log is shallow. For SOX-track companies (rare at our stage but real for a few), this becomes an issue at audit. Above $50M revenue with a public-company trajectory, plan the migration off.
Revenue recognition. QBO does not handle complex revenue recognition (ASC 606 with multiple performance obligations, deferred revenue scheduling, contract modifications) without bolted-on tools or manual workarounds. For most SaaS companies at this stage the manual workaround is acceptable through Series B.
Reporting. QBO's reports are functional but visually rough. Real board-ready reporting usually requires exporting to Google Sheets, layering Rillet on top, or generating decks separately.
Pricing at scale
Edge cases
Rillet on top of QBO. Once you cross $5M ARR or want a faster, cleaner monthly close, layer Rillet on top of QBO. Rillet pulls from QBO, automates accruals, generates close packages, and produces board-ready reporting. About $1,000 to $2,500 per month. We use Rillet on roughly half of post-Series-A engagements. It does not replace QBO; it sits on top.
When NetSuite is the right answer. NetSuite is correct when you have multi-entity (especially international), strong audit pressure (Series B+ with audit-track CFO), or revenue complexity that QBO truly cannot handle (consumption-based pricing with complex billing rules, large multi-element arrangements). First-year cost is $50,000+ and implementation is 4 to 6 months. For 95% of seed-to-Series-B companies, this is premature and a $200,000 mistake.
Sage Intacct as the dark horse. Sage Intacct is the right answer for a narrow band of companies: $20-50M revenue, multi-entity domestic, no need for international. Cheaper than NetSuite and stronger than QBO at the top end of QBO's range. We rarely recommend it because the band where it is the right answer is narrow.
Who QBO is wrong for
Multi-entity international operations (NetSuite or Sage Intacct). SOX-track public-bound companies past Series B (NetSuite). Companies with serious inventory or revenue recognition complexity that the bolt-on ecosystem cannot handle.
Headroom verdict
The boring answer is the right answer. QBO Online for accounting through Series B. Rillet on top once you cross $5M ARR. Resist NetSuite migration pressure until the company truly cannot operate on QBO; that day comes later than your audit-track CFO will tell you.
Equity admin and cap table
Pulley
The pick: Pulley for cap table, equity admin, 409A valuations, and option grant workflows if you are starting fresh or are early enough on Carta to make the migration worth it.
What Pulley does well
Interface. The cap table view is the cleanest in the category. Modeling dilution, simulating secondary, building waterfall scenarios, and reviewing option pool projections all happen in a single view that does not require an analyst to operate. The founder can run their own modeling without exporting to Excel.
409A turnaround. Pulley's 409A valuations come back in 7 to 10 business days consistently. Carta's has slipped to 4 to 6 weeks at peak periods, which blocks option grant issuance and creates real downstream problems.
Pricing transparency. Pulley publishes pricing on their site. Two-tier model: free for very early-stage (under $5M raised), then $1,200 to $3,500 per year scaled by stage. No mystery quote-based pricing, no surprise renewals at three to five times the prior year.
Customer service. Email goes to humans who respond in hours, not days. They do not route every question to a salesperson. The contrast with Carta on this dimension is sharp.
Founder-side advocacy. Pulley positions explicitly as "we work for founders, not investors" and the product reflects it. Investor-side features (LP reporting, fund admin) exist but are not the priority. Founder-side features (cap table modeling, option grant workflows, scenario planning) are the priority.
Where Pulley falls short
Investor reach. Most institutional investors are still on Carta on the investor side. If your investors strongly prefer to see your cap table in Carta, there is a real but small social cost to going against them. We have seen this matter exactly twice in 50+ engagements, and in both cases the cost was a 30-minute conversation, not a deal issue.
Fund admin. Pulley does not do fund admin. If your investors need fund admin services, that is a Carta or Aumni or Juniper Square problem, not a cap table tool problem. Stay focused on what Pulley actually does.
Brand recognition with senior hires. Some senior hires expect to see "their equity in Carta" because that is where their previous companies were. Purely a UX expectation; Pulley delivers identical information. Worth noting because it comes up in offer conversations, not because it is a real problem.
Pricing at scale
Edge cases
AngelList Stack. AngelList has launched a cap table product. We have not yet recommended it for any Headroom client because the integration ecosystem is not mature and the product is still rough. Worth watching, not yet recommending.
Early-stage spreadsheet pre-priced-round. If you have not raised a priced round and have under 10 grants, a clean spreadsheet is acceptable. The day a priced round closes, migrate to Pulley. Do not wait six months to "do it later." Audit trail starts the day investors come on the cap table.
Who Pulley is wrong for
Companies whose lead investor will absolutely insist on Carta and whose founder does not want the conversation. Companies with multi-class share structures or international entity holdings that exceed Pulley's coverage (rare at this stage). Companies with 50+ existing grants on Carta where the migration cost outweighs the benefit.
Headroom verdict
The recommendation if you are starting fresh. The migration target if you are early enough on Carta to make the switch cheap. The non-recommendation if you have hundreds of grants on Carta already; sunk cost is real and the cap table integrity at Carta is reliable.
Equity admin (incumbent)
Carta
The pick: Stay on Carta if you are already there with material grant volume. Move to Pulley if you are early. Either way: use the cap table, ignore the rest of the suite, and treat the relationship as a utility.
We have spent more time wrestling with Carta than with any other tool in this stack. The product itself is competent for its core function. Everything around it has degraded. The honest assessment matters because most clients arrive at Headroom already on Carta and the question is what to do about it.
What Carta does well
Cap table integrity. The cap table itself, as a record of who owns what, is reliable. We have not seen Carta lose data, miscalculate dilution, or misissue grants in any client engagement. The core product works.
Board consents. The workflow for board consents on grants and other corporate actions is mature. DocuSign-equivalent execution, audit trail, version history. This is the one operational area where Carta still leads.
Investor familiarity. Most institutional investors are familiar with the Carta interface for portfolio cap table review. There is a small smoothness benefit when investors do their own quarterly checks.
Brand recognition with employees. Senior hires have often seen Carta before. The "where is my equity" question has a familiar answer.
Where Carta falls short
409A turnaround. Has slipped to 4 to 6 weeks at peak periods, which blocks option grant issuance for new hires. Pulley's 7 to 10 business day cycle is a meaningful operational advantage.
Customer service. Materially degraded since 2023. Email response times of 5 to 10 business days for non-urgent issues. Phone routing to international queues with limited authority to resolve issues.
Upsell pressure. Aggressive sales motion. Expect monthly outreach pushing fund admin, equity execution services, secondary marketplace participation, valuations consulting, and CPA partnership programs. The pressure is unrelenting and the right response is to ignore it.
Fund admin and equity execution quality. Both are downstream products that have gotten worse. Multiple Headroom clients have moved fund admin off Carta in the last 18 months because of execution issues. Use Aumni or Juniper Square if fund admin is the requirement.
Pricing opacity. Quote-based pricing means the price you pay depends on the deal you negotiate. We have seen 2 to 3x variance for similar company profiles. Always negotiate at renewal; their first quote is not the floor.
Pricing at scale
Edge cases
When to migrate off Carta. If you have under 10 grants and are pre-Series A, the migration cost to Pulley is hours, not weeks. Do it. If you have 50+ grants, multiple option pools, and complex secondary history, migration cost is 4 to 6 weeks of internal time and the value of migration is marginal. Stay.
When to consolidate Carta usage. Even if you stay on Carta for the cap table, audit what else you are paying them for. Most companies have inadvertently bought equity execution services, 409A subscriptions, cap table strategy consulting, and other modules they do not use. The annual audit usually finds 30 to 50 percent of Carta spend that can be cut without operational impact.
Negotiating renewal. Always have a competing Pulley quote in hand at renewal. The mere existence of the alternative typically reduces the renewal quote by 20 to 40 percent. Their account team has discretion they will not volunteer.
Who Carta is wrong for
Pre-Series A companies starting fresh (use Pulley). Companies with founders who hate aggressive sales motion (the upsell pressure is constant and tiring). Companies who need fund admin or equity execution services (use Aumni, Juniper Square, or specialized providers).
Headroom verdict
Use the cap table, ignore the rest, push back on every upsell. Migrate to Pulley if you are early enough that migration is cheap. The relationship is a utility, not a partner; treat it accordingly.
Sales tax compliance
Anrok
The pick: Anrok for SaaS sales tax compliance. Avalara if you sell physical product. The category needs a tool; the spreadsheet approach guarantees a surprise tax bill within 18 months.
What Anrok does well
SaaS-native design. Anrok was built for SaaS companies from day one. The nexus monitoring, exemption certificate handling, invoice tax application, and product-level taxability rules are all designed around SaaS-specific complexity (states tax SaaS differently, and the rules change quarterly).
Stripe integration. Native sync with Stripe for transaction-level tax monitoring and application. Most other tools require middleware or manual reconciliation. The integration is one click and reliable.
Nexus monitoring. Anrok continuously monitors revenue and transaction thresholds across all 50 states and proactively flags when you are approaching nexus. Saves the "we found out about this state when they sent us a $40,000 bill" surprise that hits roughly half of post-Series-A SaaS companies.
Registration handling. Anrok will register your entity in any state where you cross threshold, file the returns, remit the tax. End-to-end, you do not touch the state portals. This is where the time savings come from; state sales tax portals are uniformly hostile.
Customer service quality. Responsive and competent. Their team has actual tax expertise, not just product knowledge. When you have a question about a specific state's treatment of a specific feature, they can answer it.
Where Anrok falls short
US-only. Anrok handles US state and local sales tax. For international VAT and GST (EU, UK, Australia, Canada), you need a separate tool. Quaderno for European VAT, Stripe Tax for the broader international set, or a local-specific solution per jurisdiction.
Physical product. Designed for SaaS. If you sell physical product with SKU-level tax complexity, Avalara is the correct answer. Anrok will technically work but you are using the wrong tool.
Pricing scales with revenue. Anrok's pricing is revenue-based, not flat. At scale ($10M+ ARR with multi-state nexus) the bill can hit $50,000+ per year. Worth it because the alternative is hiring someone, but worth knowing as you model the cost.
Pricing at scale
Edge cases
Stripe Tax as the alternative. Stripe Tax is built into the Stripe checkout flow, applies tax at transaction time, and remits returns. Cheaper than Anrok at small scale and more limited at every scale. Does not handle exemption certificates well, does not integrate with subscriptions outside Stripe, and the nexus monitoring is less proactive. For a very early-stage SaaS company under $500K ARR, Stripe Tax is fine. Past that, switch to Anrok.
International VAT. For European VAT, Quaderno is the cleanest add-on. For UK + EU + Australia + Canada GST, Stripe Tax has improved enough that it can handle the international side while you run Anrok for US. We recommend the dual-tool setup over forcing one tool to do both poorly.
When to set up Anrok. First $50,000 of ARR or first multi-state customer, whichever comes first. Setting up earlier wastes a few months of subscription on monitoring you do not yet need. Setting up later means you may already be in violation when you start.
Who Anrok is wrong for
Physical product companies (use Avalara). Pre-revenue companies with no transactions (set up at first $50K of ARR or first multi-state customer, not before). Companies with international-only revenue and no US obligations.
Headroom verdict
Default for SaaS. Set up at first $50,000 of ARR or whenever you cross your second state of customers, whichever comes first. The cost of setting up six months early is small; the cost of setting up six months late compounds monthly.
Decision frameworks
Three named frameworks to make the calls in 10 minutes
The tool sections above explain why each pick is right. The three frameworks below are how we actually run the conversation with a founder in a 30-minute discovery call. Score yourself in each one. The answer falls out.
Framework 1
The Headroom Payroll Decision
Four questions. Answer yes to all four for Gusto. Any no sends you to Rippling. International contractors get Deel in parallel either way.
Rule: Yes to all four → Gusto. Any no → Rippling. International contractors → add Deel regardless.
Framework 2
The Headroom Spend Stack Decision
Three questions. Default falls out as Ramp 80 percent of the time. The exceptions are real but narrow.
Rule: Default to Ramp. Override only on Q1 or Q2. Issue cards day one.
Framework 3
The Headroom Cap Table Decision
Three questions. Migration math is the whole framework: under 10 grants is cheap to move, over 50 is expensive, in between is judgment.
Rule: Default to Pulley unless you are deep on Carta. The 10-to-50-grant middle is judgment; lean toward staying put unless you have a strong reason to migrate.
Days 1 to 15
Money flows. Get them clean.
Operating account: Mercury
Open it within the first week of close. Move the closing wire from your investors there if it is not already. Set up two sub-accounts immediately: one for payroll funding, one for vendor and ops spend. Activate Mercury Treasury for cash you will not touch in the next 30 days. The yield matters at $5M+.
Do not run the operating account through a traditional bank. Chase, BofA, and Wells will eat 3 to 6 hours of founder time per month in friction. The wire request that takes 30 seconds in Mercury takes 90 minutes and a phone call at the legacy banks. You closed a round to spend money on growth, not on hold music.
Corporate cards: Ramp
Issue cards to every employee on day one. Set the policy at issuance: software, travel, and T&E limits encoded in the card itself, not a Notion doc nobody reads. The point is not the limits. The point is that policy lives in the tool, enforced at swipe.
Brex is fine if you have a Brex Cash account already and want one vendor. The reason most clients land on Ramp is the policy engine and the lack of upsell pressure. Brex spent the last two years pivoting toward enterprise; Ramp stayed focused on the post-raise stage you are in.
Avoid AmEx for company spend. The points are nice, the policy controls are 1995-era. If you must keep an AmEx open for vendor compatibility, fine, but it is not your spend tool.
Payroll: pick one and commit
If you have under fifty W-2 employees in three or fewer states, no global team, and no equity comp complexity: Gusto. Set it up in 48 hours. The UI is good enough that you can run it yourself for the first six months without a payroll specialist.
If you have employees in five or more states, plan to hire internationally, or want HRIS and payroll in one surface: Rippling. The setup is heavier (give it two weeks), the price is higher, and the lock-in is real. The tradeoff is one source of truth for headcount, payroll, devices, and access.
Do not pick ADP or Paychex. The pricing, the UI, and the integration story are all built for companies five times your size. Do not pick a "fractional payroll" service that runs Gusto for you. You will pay $1,200 a month for them to push the same buttons you would.
83(b) elections
If anyone joined in the last 30 days and received early-exercisable stock, their 83(b) window is closing. Put a calendar reminder, chase down certified mail receipts, and confirm postmark dates. A missed 83(b) costs the employee personally, which means it eventually costs you in retention and trust.
Multi-state payroll registrations
The day you onboarded a remote employee in a new state, you took on registration obligations there. Rippling and Gusto will handle most of the registrations themselves, but the application can take 4 to 8 weeks per state. Start the slow ones now: California, New York, Massachusetts, Washington, Pennsylvania.
If you are behind on registrations, Mosey is a separate tool that does nothing but track and complete state filings. Worth the $400 per month for the next two quarters until you are caught up.
Days 15 to 45
Books and equity. Build for the next audit, not the last one.
Accounting: QuickBooks Online
The decision is boring and the answer is the same for 95 percent of post-raise companies through Series B. QBO Online. Not Desktop. Not NetSuite.
Do not move to NetSuite at this stage. NetSuite is the right answer eventually, usually after $30M ARR, multi-entity, or strong board pressure from an audit-track CFO. At Series A it costs you $50,000+ in first-year fees, six months of implementation, and a bookkeeper who will resist the migration when you have to do it again at scale. QBO will hold you up to and through Series B for most companies.
Do not stay on QuickBooks Desktop. If you inherited it from your accountant or your last job, migrate. The data export and reimport into QBO Online takes a weekend.
Chart of accounts
The default QBO COA is wrong for a SaaS or AI company. Customize it before you have transactions in it. Specifically: a separate revenue account for each product line, a cost-of-revenue section that captures hosting, AI inference costs, and direct customer support, and operating expense buckets that match how you will explain the business to a board. Once you have booked 90 days of transactions against a wrong COA, the cleanup is painful and the historicals never look quite right again.
Monthly close
Target Day 5. That means by the fifth business day of every month, last month's books are closed, the P&L and balance sheet are in your inbox, and there is a one-page variance memo for anything that moved more than 10 percent. If you cannot hit Day 5 yet, target Day 10 for the first three months and then tighten. Day 15+ closes are common at this stage and they are a red flag, not a phase.
The close requires four things to function: a documented checklist, accruals booked the same way every month (rent, software, payroll), bank reconciliations that match the Mercury and Ramp feeds, and a single person who owns the calendar. The single owner is the part most companies miss. "Everyone owns it" means no one does, and Day 5 becomes Day 22.
Equity admin: Pulley first, Carta only if forced
If you do not have a cap table tool yet, set up Pulley. The UI is better than Carta's, the pricing is honest, and customer service responds to email instead of routing you to a salesperson. Pulley handles 409A valuations, option grants, board consents, and shareholder records cleanly through Series B and beyond.
If you are already on Carta, stay there. The migration cost is real, and the parts of Carta that work (cap table integrity, board consent infrastructure) are reliable enough. Use Carta for the cap table and ignore the rest of the suite. Their fund admin, equity execution scheduling, and 409A turnaround times have all gotten worse in the last two years, and the upsell pressure is constant. Pull what you need, push back on the rest.
409A within 60 days of close
Whichever platform you are on, get a 409A scheduled within 60 days of your priced round. Strike prices on options issued before the 409A is dated are exposed. Do not issue option grants in the gap. If you have hires waiting on grants, hold them until the valuation is in.
Quarterly grant cadence
Once a quarter, all new hires receive grants on the same date and board consents go out together. Doing this monthly means twelve board consent rounds a year and twelve chances to forget one. Quarterly is the sweet spot until you are 100+ employees and need to tighten further.
Vendor consolidation
By Day 45 you should know exactly what software the company pays for. Ramp's vendor view will surface this automatically once cards are in use. Cancel the duplicate Notion accounts, the trial subscriptions someone forgot, the seven Loom seats nobody is using. Most post-raise companies waste 15 to 25 percent of software spend on this. The audit takes two hours and pays for itself the same week.
Days 45 to 90
Compliance, controls, and the handoff
Sales tax nexus
If you sell SaaS, you have nexus exposure in any state where you cross a customer threshold (usually $100,000 in revenue or 200 transactions). Most post-raise companies are over the threshold in 5 to 10 states without realizing it. The penalty for non-collection compounds monthly, and back-collection from customers is rarely possible.
Anrok. Hook it up to Stripe and your CRM, let it monitor nexus across states, register where required, file returns. The setup is one afternoon and the ongoing monitoring is automatic. There is no version of this you handle with a spreadsheet at scale.
If you sell physical product, Avalara is the equivalent. Different tool, same logic.
1099 contractor compliance
Any contractor you paid $600 or more to in the calendar year gets a 1099-NEC by January 31 of the following year. If your contractors are paid through Deel or Rippling, the platforms handle the filing. If they are paid through Stripe, Wise, or direct deposit out of Mercury, you are on the hook to track and file. Build the contractor list now, not in December.
Expense policy, made real
Document it, share it, enforce it through Ramp. The policy lives in three places: the Ramp policy engine, a one-page Notion doc all employees acknowledge in onboarding, and the offer letter for new hires. Three places, same content, no contradictions. Inconsistencies between the three are how disputes start.
Approval workflows
Anything over $1,000 needs founder or finance approval before purchase, not after. Ramp, Mercury wires, and your accounting close all need to reflect the same approval thresholds. Inconsistencies here are how companies end up with $30,000 of unbudgeted spend in a quarter and no one is sure who approved it.
Board reporting cadence
By Day 75 you should have a monthly financial pack that goes to the board: P&L vs. budget, balance sheet, cash position with a runway calculation, and a one-page CEO commentary. Not a slide deck. A document. Slides are for the meeting; the document is for the board members who actually read.
Insurance review
Most post-raise companies discover their coverage is wrong about the time someone files a claim. By Day 75, run a current policy review with your broker: D&O, E&O, cyber, employment practices liability, and general commercial. The conversation is 30 minutes and the gaps are usually significant. Post-priced-round D&O coverage in particular tends to be undersized for the new cap structure.
DE franchise tax and foreign qualifications
If you incorporated in Delaware (most do), the franchise tax bill comes due March 1 every year. Set the calendar reminder now. The Assumed Par Value method almost always saves you several thousand dollars over the default Authorized Shares method; make sure your accountant is filing it correctly.
For every state where you have employees or material revenue, confirm you are foreign-qualified to do business there. Most companies miss two or three of these and pick up small but compounding penalties.
The runbook
By Day 90, every system above has a written runbook: who owns it, what runs through it monthly, what the failure modes are, and where the credentials live. The runbook lives in a single Notion or Google Drive folder, not scattered across Slack DMs and someone's laptop. One folder. One source of truth.
This is the artifact you hand to whoever you eventually hire. With it, your first finance or ops person ramps in three weeks instead of three months. Without it, they spend their first quarter reverse-engineering what you built, which is the most expensive way to get a system documented.
If you do not want to build the runbook yourself, this is exactly what Headroom delivers at the end of a sprint.
The checklist
Track your 90 days
Below is the same checklist we run for Headroom clients. Tick items off as you complete them. Your progress is saved locally in your browser, so you can come back to it across the 90-day window without an account or a login.
If anything in here does not make sense for your company, skip it. If everything in here does not make sense, that is the gap Headroom fills. The checklist is opinionated by design; use it as a starting point, modify it where your situation requires, and do not add items just because someone on Twitter said they were important.
90-day back office
Your build progress
Phase 1 / Days 1 to 15
Money flows
Phase 2 / Days 15 to 45
Books and equity
Phase 3 / Days 45 to 90
Compliance, controls, handoff
Scenarios appendix
When your company is not the standard case
The default playbook above assumes a US-only SaaS company at seed or Series A. That covers roughly 60 percent of post-raise founders. The five scenarios below cover the most common deviations and what changes in each. If your company looks like one of these, treat the relevant section as an override layer on the default stack.
Scenario 1
25-person SaaS with a UK or Canadian entity
The single biggest stack change vs. the default: payroll. Your US-only Gusto setup does not work. Either run Gusto US plus Rippling Global UK in parallel, or run Rippling for both jurisdictions. Rippling is the cleaner answer past five UK employees because the data model is shared and the payroll runs together.
Banking splits. Mercury does not hold GBP or CAD balances. Add Wise Business or Revolut Business for the UK, Wise or RBC Royal Bank for Canada. Mercury for the US entity, Wise for the foreign entity, and a documented intercompany funding flow between them with quarterly true-up.
Sales tax becomes more complex. Anrok handles US. For UK VAT, add Stripe Tax or Quaderno; the two run in parallel, neither replaces the other.
Equity comp for UK employees should use EMI options if eligible, not US ISOs and NSOs. EMI has favorable UK tax treatment but requires HMRC valuation and proper grant documentation. Pulley and Carta both handle EMI; verify scheme compatibility with your specific structure before granting.
Estimated stack cost premium vs. US-only: $2,500 to $4,000 per month additional, roughly half from payroll and half from compliance.
Scenario 2
AI/ML company with heavy compute spend
The COGS structure is different and your QBO chart of accounts needs to reflect it. Separate accounts for AI inference costs (OpenAI, Anthropic, Bedrock, Vertex), training compute (CoreWeave, Lambda, AWS H100 instances), and infrastructure hosting (AWS, GCP, Azure non-AI services). Most accountants will lump these into "hosting" and you will lose all unit economic visibility.
Gross margin reporting becomes load-bearing. If you report a 70 percent gross margin to your board with all compute costs in COGS, that is the right number. Hiding compute in operating expenses misleads the board about your unit economics. The right COA prevents this from being an argument later.
Vendor consolidation matters more. Track every AI vendor by spend, by use case, and by who in the company is using them. Ramp surfaces this; the audit reveals 20 to 40 percent waste in our typical engagement, often more for AI-native companies running parallel evals on multiple model providers.
Cash management discipline tightens. AI companies burn cash 2 to 3x faster than non-AI SaaS at the same revenue level. Mercury Treasury yield matters more, runway calculation matters more, monthly burn variance matters more. Tighten the close to Day 5 and run variance reviews bi-weekly rather than monthly.
R&D tax credit eligibility: most engineering and ML work qualifies for the federal R&D credit and state versions in California, New York, and Massachusetts. Engaging a credit specialist (Ardius, Strike Tax, or Gusto's R&D credit feature) at year one is worth $50,000 to $300,000 depending on burn and headcount. Often pays for the full Headroom engagement.
Scenario 3
Marketplace with payouts to users or sellers
The money movement is fundamentally different. You are running both AR (taking money from customers) and AP-at-scale (paying users). The volume changes the toolset.
Stripe Connect or Stripe Treasury is usually the answer. The platform handles payout flow, 1099-K filings (over $600 per user per year), and KYC. Trying to handle this through Mercury direct disbursement breaks at scale, both operationally and from a compliance perspective.
Reserves matter. Most marketplaces hold reserves against chargebacks, refunds, and platform liability. Your accounting needs a reserve liability account, properly accrued. Not a default QBO COA setup; build it explicitly.
Tax 1099-K filings hit thousands of users annually. Stripe Connect handles the filing. Anrok still handles the marketplace's own sales tax obligations. The two are separate and both required.
State money transmission licensing is a real risk. Some marketplace structures, especially those holding user funds for any duration, trigger money transmitter license requirements in 30+ states. Get a fintech attorney to opine on your specific structure before you scale. This is a category where bad legal advice costs millions.
Scenario 4
Hardware company at Series A
Inventory accounting is non-negotiable. QBO's basic inventory is not enough. Layer Cin7 or Katana on top for SKU-level tracking, COGS calculation, and warehouse management. The integration with QBO is mature; do not try to run inventory inside QBO.
Contract manufacturer payments. Most CM relationships use international wires denominated in USD. Mercury handles outbound USD wires globally. If you are paying a CM in CNY or JPY, Wise Business in parallel is cheaper than Mercury's FX margin by 0.5 to 1.5 percent per wire, which adds up at scale.
Parts and components inventory creates working capital tightness that pure SaaS founders do not anticipate. Cash flow forecasting becomes a weekly exercise, not monthly. Layer in 13-week cash flow modeling at Day 30 of the playbook, not Day 75.
Sales tax for physical product: Avalara, not Anrok. Different calculation engine, different exemption certificate handling, different state filing logic. Do not try to make Anrok work for hardware; use the right tool.
Customs and import duties: not a back-office tool, it is a customs broker relationship. Flexport, Freightos, or a traditional broker. Most hardware founders have this from day one; if you do not, get one before your first international shipment.
Scenario 5
Dev tools company with distributed engineering
Most of your team is international, most of your customers are US. The stack splits asymmetrically: international payroll and contractors heavy, US-light operationally.
Default stack: Rippling (or Deel) for global engineering, Gusto or Rippling for US-side founders and execs, Mercury US for the operating account, Wise or Revolut for international payouts where they are not handled inside the payroll platform.
IP assignment is critical. Engineers in Germany, Brazil, Russia, and several other jurisdictions have automatic IP carve-outs in default contractor agreements. Use Deel's jurisdiction-specific contractor templates, or have local counsel review per jurisdiction if you write your own. Generic US-form contractor agreements are unenforceable in many of the jurisdictions where you will hire.
Equity for international hires: see the EMI note in Scenario 1, applied across jurisdictions. Most countries have specific tax-favored equity programs. Default US ISO/NSO grants are usually the worst tax outcome for non-US employees. Pulley and Carta both handle major variants; design grants per jurisdiction, not per US default.
Time-zone-driven close. If your books close on US time but your engineering payroll runs on Berlin time, build the close calendar with the lag in mind. Day 5 close target shifts to Day 7 or Day 8 to allow for international payroll cutoff and currency conversion timing.
Avoid these
The traps founders fall into
A few things show up post-raise that look reasonable in the moment and cost real money later. Skip these.
- NetSuite at Series A. The $50,000+ first-year cost and six-month implementation buy you nothing you cannot get from QBO. Wait until $30M+ ARR or strong audit pressure forces it.
- Bill.com as the AP backbone. The UX is a museum piece, the integration story is weak, and the price is wrong. Use Ramp Bill Pay or Mercury for AP. Bill.com migrations away are a recurring Headroom engagement on their own.
- ADP or Paychex. Built for companies five times your size. The UI will frustrate you and the implementation cost will surprise you. Gusto and Rippling are the right answers at this stage.
- A spreadsheet cap table after a priced round. The day you have outside investors, you have audit trail obligations. Pulley or Carta is the answer; the spreadsheet is a liability.
- "Fractional CFO" for build-out work. Different category. Fractional CFOs do strategy and board narrative. The build (payroll, close, spend, equity, compliance) is operations work. Get the build done first, layer in a fractional CFO later if you need one.
- A Head of People before the systems exist. The most common mis-hire post-Series A. They show up to a company with no payroll runbook, no benefits enrollment, no equity admin, and end up doing the work themselves while the title burns.
- Letting state payroll registrations slide. Each state you owe in adds penalties monthly. A "we will get to it" answer at month two is a $20,000 surprise at month eight.
- Carta upsells. Use the cap table, ignore the rest. Their fund admin, valuation scheduling, and equity execution have all gotten worse, and the sales pressure is unrelenting. Treat the relationship like a utility, not a partner.
Founder FAQ
Questions we hear
Can a founder run this themselves?
Most can run 70 to 80 percent of it. The parts that consistently break under founder ownership are the monthly close (because no one owns it after Day 5), multi-state payroll registrations (because the workflow is opaque), and sales tax nexus monitoring (because no one notices until a state sends a bill). If those three are handled, founder-run works.
What does the recommended stack cost in software fees?
For a 25-person post-raise company, expect $4,000 to $7,000 per month across Mercury (free), Ramp (free for cards; paid for Bill Pay), Gusto or Rippling (~$60 to $100 per employee per month), QuickBooks Online (~$90 per month), Pulley or Carta ($1,500 to $4,000 per year), Anrok ($500 to $1,500 per month), and 1Password ($8 per user per month). Payroll is the biggest variable and scales with headcount.
What if we are already past 90 days post-close?
Run the playbook anyway. The order matters less than the completeness. Most companies that hit Day 180 with gaps still benefit from running the same sequence. It just takes longer because you are undoing prior decisions while building new ones.
Should we hire a fractional CFO instead of running this?
Different problem. A fractional CFO produces analysis, board narrative, and forecast modeling. None of that requires payroll to be set up correctly. The build needs to happen either way. If you want both, do the build first, then layer in a fractional CFO once the systems are running.
Why not just hire a controller?
A controller without a system to control spends their first quarter building one. That is a $50,000 to $80,000 engagement cost wrapped in a permanent salary, with no defined end date. Build the system first; then a controller can step into something they can run.
What does it cost to get the first 90 days wrong?
A bad first 90 days typically costs $150,000 to $400,000 in the first year: penalties for missed state registrations, mis-hires built on top of a missing foundation, software you will migrate off of, and founder hours diverted from product and GTM.
How does the stack change for international hires?
Two changes. First, payroll: add Deel for international contractors or Rippling Global for international employees. Gusto does not run international payroll. Second, banking: Mercury does not hold non-USD balances, so add Wise Business or Revolut Business in parallel for any foreign entity collecting GBP, EUR, or CAD. Equity grants for international employees should use jurisdiction-specific programs (EMI in the UK, qualifying options in France) rather than default US ISOs and NSOs. See Scenario 1 for the full UK and Canadian playbook.
Should we use Ramp or Brex?
Ramp 8 times out of 10. The exceptions are real but narrow: companies committed to the Brex Cash + cards stack, companies with significant non-USD international card issuance needs, and companies with a strong existing Brex relationship and active line of credit. The reason most clients land on Ramp is the policy engine, the QBO sync depth, the free-for-life pricing on the core product, and the lack of upsell pressure. Brex is competitive on the card itself; the surrounding product has been pulling toward enterprise.
When should we move from QuickBooks Online to NetSuite?
Later than your audit-track CFO will tell you. NetSuite is correct when you have multi-entity (especially international), strong audit pressure (Series B+ on a public-company trajectory), or revenue complexity that QBO cannot handle (consumption-based pricing with complex billing rules, large multi-element ASC 606 arrangements). For most companies that day comes after $30M ARR. First-year NetSuite cost is $50,000+ in fees and 4 to 6 months of implementation. Migrating prematurely is a $200,000 mistake. If you need more close discipline before then, layer Rillet on top of QBO.
Keep reading
More on how we work
Get in touch
Tell us where things stand
If you just raised and your back office has not kept up, email hello@headroom.llc or use the form on the home page. We will follow up within a couple of days.