The Founder's Checklist for Launching an Online Store

Most online store launches don't fail because the product was wrong. They fail because of something small and boring: a payment gateway that was never tested with a real card, a tax setting left at the default, a checkout button that's unreachable on an iPhone. None of these show up when you're staring at your homepage on a laptop. All of them show up the moment a real customer tries to give you money.

This checklist is built for the founder who is not a developer and is not going to become one before launch day. It's organized by how close you are to going live, not alphabetically — because the order you check things in matters just as much as whether you check them at all.

The 60-Second Version

If you only have time to skim, these are the seven items that cause the most launch-day damage when skipped:

• Place a real test purchase with a real card, start to finish
• Confirm tax is calculating correctly for every region you sell into
• Confirm shipping rates and delivery zones are set up, not left on defaults
• Complete a full checkout on an actual phone, not just a resized browser window
• Check that your page loads in under 3 seconds on mobile
• Publish Terms of Service, Privacy Policy, and a clear Returns policy
• Confirm your analytics tool is receiving real events, not just installed

Now the full breakdown, grouped by how much time you have left.

Three to Four Weeks Out: Get the Money Plumbing Right

These are the items that take the longest to fix if something's wrong, so they need to be locked in early.

Test your payment gateway with real transactions

Most store owners test payments once, in "test mode," during setup, and never touch it again. That's not enough. A week or two before launch, switch to live mode and run a real, small purchase with an actual card. Then refund it. You're checking three things: the money actually moves, the confirmation email fires, and a refund doesn't get stuck in limbo. If you accept more than one payment method — card, PayPal, Apple Pay, buy-now-pay-later — test each one separately. They fail independently.

Set up tax correctly for every region you sell to

This is the item non-technical founders skip most often, and it's the one that creates real legal and financial exposure. Sales tax in the US varies by state and sometimes by city. VAT in the EU applies differently depending on where your customer is and what you're selling. Your ecommerce platform almost certainly has a tax module — the mistake is assuming it configures itself. Confirm your tax rules are switched on for every region you actually ship to, not just your home market.

Configure shipping zones and rates deliberately

"Free shipping everywhere" sounds generous until you get an order from somewhere your margins can't absorb. Before launch, decide explicitly which regions you ship to, what each zone costs, and what happens with oversized or heavy items. Then place a test order from an address outside your home country if you sell internationally, to confirm the right rate actually appears at checkout.

Two Weeks Out: Make the Store Work the Way Customers Actually Shop

Run the entire checkout flow on a real phone

Most ecommerce traffic arrives on mobile, and most launch bugs live in mobile checkout — a payment button hidden below the fold, a form field that doesn't trigger the right keyboard, an address field that's too narrow to read. Testing on a laptop with a resized browser window doesn't catch these. Borrow a phone, or use your own, and buy something from your own store, thumb to thumb, the way a customer would at 11pm on the couch.

Check your page speed, not just your homepage

A slow product page or a slow checkout page loses sales before a customer ever sees your price. Run your product pages and your cart through a free speed-testing tool and look at mobile results specifically, since mobile connections are slower than what you're testing on in your office. Oversized product images are the most common cause of a sluggish store, and usually the easiest to fix.

Publish your legal pages before you publish your storefront

Terms of Service, a Privacy Policy, and a clearly stated Returns and Refunds policy aren't paperwork — they're conversion tools. Customers hesitate to buy from a store that doesn't visibly stand behind its products, and in most regions where you'll sell, some version of these pages is a legal requirement, not a nice-to-have. If you take payments from EU customers, you also need a cookie consent banner that lets people actually decline non-essential cookies, not just acknowledge a message.

Launch Week: What Founders Forget Under Time Pressure

By now the big structural pieces are in place. This is where the details that get rushed live.

Place one more real order, end to end

Do this again, even though you did it three weeks ago. Platforms update, plugins update, settings drift. A final real purchase — card charged, confirmation received, order visible in your admin panel — is the single highest-value five minutes you'll spend in launch week.

Confirm your analytics are actually receiving data

Installing an analytics tool and confirming it's working are two different things. Open your analytics dashboard, browse your own store in another tab, and watch for your own activity to appear. If nothing shows up, you'll launch with zero visibility into what your first real customers are doing — and you won't find out until it's too late to fix it cheaply.

Test every automated email a customer will receive

Order confirmation, shipping notification, password reset, abandoned cart reminder — each of these is a separate email template, and each one can independently break. Trigger every one of them and check that they land in an inbox, not a spam folder, and that every link inside actually goes where it should.

Check what happens when something goes wrong

Add an out-of-stock item to a cart. Try an invalid discount code. Enter a card number that will get declined. A store that only works when everything goes right isn't ready — customers who hit an error with no clear next step usually just leave, and some of them tell others why.

24 Hours Before You Flip the Switch

Do one last pass with fresh eyes, ideally someone who hasn't looked at the store in the past week — a co-founder, a friend, anyone who will notice what you've stopped seeing. Confirm the site loads for a first-time visitor with no account, no saved payment details, and no prior context. Decide, in writing, who is checking orders and answering customer messages in the first 48 hours, and how they'll reach you if something breaks.

Launch Day and the Week After

Traffic on launch day tells you less than you'd think. What matters is the conversion funnel: how many visitors add something to a cart, how many start checkout, and how many finish. A gap at any one of those steps points you directly at what to fix next, which is far more useful than watching a visitor counter climb. Keep customer support visible and responsive during this window — the first customers of any new store are also its most forgiving, but only if problems get resolved quickly and honestly.

Getting Launch-Ready With Help

This checklist covers what tends to break. It doesn't cover platform selection, which is its own decision with real trade-offs between cost, flexibility, and how much ongoing technical work you're signing up for — something we cover in our comparison of ecommerce platforms for startups.

At P2C, we build and launch online stores for founders who don't want to become ecommerce platform experts to run their own business. If you'd rather have a second set of eyes go through this checklist with you before you go live, we're glad to help.

Key Takeaways:

• Order by urgency, not alphabet — payment and tax setup need weeks of lead time, final checks need days
• A real purchase with a real card is the single most reliable launch test you can run
• Mobile checkout deserves its own dedicated test pass, on an actual phone
• Legal pages and cookie consent aren't paperwork — they affect both compliance and conversion
• Analytics that "look installed" and analytics that are actually receiving data are not the same thing
• Launch week is for verification, not new features — resist the urge to add anything untested