AtClearspace (YC W23), our mission is to help individuals forge healthier relationships with their devices. At Y Combinator, they teach startup founders the most valuable way to spend their time is by doing two things: building products and talking to users. But that isn't all there is to start a company, as co-founder and CEO ofClearspace, my job also includes making sure we steward the capital we have to pursue our mission to our fullest capacity.
At YC, they teach the Paul Graham principle do things that don't scale - which has been the secret sauce to finding product-market fit for billion-dollar startups like Stripe, Airbnb & Coinbase.
TLDR:
At Clearspace, this meant shipping an early version of our app (now available on both iOS and Android) with a 5-minute onboarding video, writing articles teaching screen time tricks likegrayscale mode for screen time, and hosting a first-timeNo Phone Socials at our office in Silicon Valley.
But doing things that don't scale is only a virtue until you start to grow. As the wheels get turning and you begin to crack the code on execution and implementation, it then becomes your job to find ways to optimize your startup ideas.
Puzzle is the tool that has supercharged our ability to intelligently manage our financials as an early-stage startup.
Monthly finance meetings before we were on Puzzle were the meetings that our team dreaded. The 1-hour Friday afternoon slot always ended up taking 2-3 hours.
We would skim through our Mercury account trying to make sense of each spend metric, copying and pasting key line items into an Excel sheet to help us calculate runway, a pain point well-documented for anyone looking at accounting software for Mercury users. The type of meeting that would have a lot of people and even the Clearspace team members tempted to pull out their phones and start doom-scrolling.
Once we got on Puzzle, we could take care of financial housekeeping in minutes instead of hours and allocate extra brain cycles to get strategic about our financial picture.
Our monthly finance meeting has become a place where my co-founder Oliver and I run through headcount scenarios and spend time using Puzzle dashboards to inform discussions about when to raise and what our burn rate and unit economics are looking like: the kind of burn rate clarity for founders that used to require a full CFO.
We leave these meetings informed and excited about what our business can do in the future instead of being drained and confused.
| Before Puzzle | After Puzzle | |
|---|---|---|
| Meeting length | 2 to 3 hours | Minutes |
| Primary activity | Copying line items into Excel to calculate runway | Running headcount scenarios and reviewing dashboards |
| Data source | Manual bank account review (Mercury) | Puzzle dashboards with burn rate and unit economics |
| Team sentiment | Drained and confused | Informed and excited |
| Strategic output | None: consumed by data wrangling | Informed decisions on when to raise and how to spend |
My background is in Software Engineering, not finance, so bringing together high-quality investor reports was a steep learning curve. This is where Puzzle feels like it gives me superpowers, especially when preparing for fundraising (see The Founder's Guide to Fundraising for what investors actually expect). Where QuickBooks contains all the data I need but doesn't present it in startup terms, which is exactly why so many founders look at QuickBooks alternatives for startups...
Puzzle actually pulls the financial picture together in a way that makes sense in the native language of startups: runway, spending, revenue, etc., paired with the types of tooling you'd expect: monthly checklists, annotations, intelligent insights that are familiar in my non-technical tool stack:Superhuman, Notion, Slack, Puzzle.
Puzzle’s generative system is already an incredible value-add for our team, but what got me fired up enough to write an unsolicited tweet is what I see coming down the pipeline from them.
You'll get it within the first 5 minutes of using the product. Puzzle is building the solution you'd hoped existed by now as a founder, and they aren’t slowing down. Across the UI/UX, you can see tells of intelligent and useful financial tools to come.
If that wasn't enough, the Puzzle team is high touch, always asking for feedback, and excited to share their story. I've been in a few pinches where I'll need a quick Puzzle tip or general startup accounting advice, and a member of the Puzzle team comes back with high-value input.
Puzzle is one of the pieces of software our company wouldn’t be the same without.
At Clearspace, we measure our impact by how much time we give back to our users. We've now crossed 25 million hours saved, and I can confidently say Puzzle has meaningfully contributed to that mission by freeing us to do what we do best.
This is one of the most common questions founders run into when they first go accrual. Puzzle maintains both cash and accrual books simultaneously, so the cash activity report reflects actual money moving in and out of your bank accounts, while the balance sheet is built on accrual-basis entries: revenue recognized when earned, expenses when incurred, not when cash changes hands. For a deeper look at why startups need both, see our guide on cash vs. accrual accounting. For accrual-basis clients, timing differences (like deferred revenue, accounts receivable, or prepaid expenses) will always cause the two to diverge. That's expected behavior, not a bug. If you see a gap you can't explain, the usual culprits are uncategorized transactions, unreconciled bank feeds, or accrual entries that haven't been posted yet. Puzzle's month-end checklist surfaces these before they compound.
Puzzle connects natively to the fintech stack most early-stage startups already run: Stripe, Mercury, Ramp, Brex, Gusto, Deel, Rippling, and more, with no middleware layer required (see the full integrations list). The Stripe accounting integration is where Puzzle earns its keep compared to QuickBooks: QuickBooks treats each Stripe payout as a single lump-sum deposit, which means your revenue recognition is off and your books don't reflect actual transaction-level data. Puzzle pulls in individual Stripe charges, refunds, fees, and payouts, maps them to the right revenue and liability accounts, and applies ASC 606-compliant recognition automatically. The result is accurate deferred revenue on your balance sheet from day one, with no manual workaround or custom spreadsheet your accountant has to maintain.
When Clearspace first got on Puzzle in 2023, we were early adopters of AI-native accounting. By late 2026, competitors like Rillet and Digits have added similar AI-native features. The AI accounting software market appears to be shifting toward tools built natively for the startup fintech stack, not legacy platforms retrofitted with AI features. Puzzle has kept pace with that shift. Since we onboarded, they've shipped one-click global payroll sync via Deel, a rebuilt integrations page with at-a-glance account health and on-demand manual sync, bulk classification rules that automate how you tag transactions across departments and projects, and task and status dashboard cards that surface what needs attention without hunting through menus. Each of these would have taken us hours of manual work per month in our pre-Puzzle workflow. The product we're using today is meaningfully more capable than the one we started with, and the team is still shipping.
When we started using Puzzle in 2023, AI-native accounting felt like a novelty. In 2026, it is the expectation. The startup finance tooling market has shifted dramatically: legacy platforms are scrambling to retrofit AI onto decades-old architecture, while purpose-built tools like Puzzle were designed from day one to do the work automatically. Real-time close and automated categorization have moved from differentiator to baseline, and Puzzle was built for exactly that shift.
For founders, that shift has a direct cost: showing up to a VC conversation without real-time financial visibility signals that your financial house isn't in order, and that perception is hard to walk back mid-round. Having infrastructure that surfaces that picture automatically isn't a nice-to-have. It's table stakes. Puzzle got us there early, and that's an advantage we're not giving back.
