Do You Actually Need an Online Store? Four Jobs People Call E-Commerce

A customer asks whether they can just give you a card over the phone. Someone suggests Amazon. The invoices are still going out by hand. So you start looking into adding e-commerce to the website, and by the end of the afternoon you've read Stripe's pricing page, a WooCommerce tutorial, and a forum thread from somebody whose payouts got frozen for three weeks.

Here's the part that saves the most money. Usually you're asking four separate questions, and only one of them is a store.

Three of those four don't need a shopping cart on your website. And anything you don't build can't break, can't get hacked, and can't need patching on a Sunday.

Taking cards over the phone

This one needs a payment processor, not a store. Stripe has a virtual terminal built into its own dashboard — log in, create a payment, key in the card while the customer is on the line. No plugin, no cart, no developer, and you can do it this afternoon.

Keyed-in cards cost more than online ones and you carry the chargeback risk, which is worth understanding before you route your biggest invoices through it. That, and the Stripe-or-PayPal question, are covered separately in Stripe or PayPal for a small business.

Sending people to your distributors

This isn't e-commerce either. It's a Where to Buy page, and it's one of the highest-return pages a manufacturer can have. For each distributor: name, territory, what they stock, phone, email, direct link. If they carry different product lines, say which — the usual failure of these pages is sending a customer to a distributor who doesn't stock what they came for.

Track the outbound clicks too, so you can tell each distributor every quarter how much traffic you sent them. That email does more for your channel relationships than a cart would.

Decide the channel question deliberately while you're at it. If you start selling direct at retail while your distributors carry your inventory, you have a conflict on your hands. Better to choose that than to trip into it because a plugin made it easy.

Amazon

Amazon is already a complete store: its own listings, checkout, payment processing and payout schedule. Money from Amazon comes from Amazon, and your own payment processor has nothing to do with it.

So "integrate our Amazon channel" means one of two very different things. Linking is a Buy on Amazon button on the product pages you already have — free, an afternoon's work, and the right answer for most manufacturers. Syncing is one shared inventory count, so an order on either side decrements the other; the official Multichannel for WooCommerce plugin does it.

Sync only earns its keep when you're selling the same stock through two channels fast enough to oversell it. Link first; sync when overselling becomes a real risk rather than a theoretical one.

Invoices and labels

This is the job most likely to be solved by something you already pay for. If you use QuickBooks or Xero, invoice from there — your accountant already reconciles it. If you don't, Stripe Invoicing sends a hosted invoice with a pay button and marks itself paid. What you don't want is invoices living in a system your bookkeeping never sees.

An invoice is also only useful if it arrives, and invoices sent by aWordPress site are among the likeliest emails to land in spam, because a server sends them rather than a person. If you're going to invoice from the site, sort that out first.

For labels, a US store billing in USD can use WooCommerce Shipping free to buy and print USPS, UPS and DHL labels at Woo's commercial rates. It needs WooCommerce installed, so it's worth it only if you're shipping enough parcels for the discount to outweigh the maintenance.

The order to do it in

1. Turn on the Stripe virtual terminal. Nothing to build, and you find out what your real volume is.
2. Build the Where to Buy page, and track the outbound clicks.
3. Add Buy on Amazon buttons to the product pages you already have.
4. Move invoicing into whatever your accountant reconciles, and fix the site's email deliverability if invoices will come from the site.
5. Only then, with three months of real numbers, decide whether a checkout is worth building.

Most of the value is in the first four, and none of them is a store.

What actually blocks this

Rarely the payment processor. Almost always the product data. Syncing to Amazon needs stable SKUs; shipping labels need weights and dimensions; a checkout needs prices in a database rather than a PDF price list. If your SKUs disagree between your website, your ERP and your Amazon listings, no plugin fixes that — it just fails more expensively later.

Taking money also raises the stakes on everything already wrong with an older site. A brochure site that's a bit slow and out of date is an inconvenience. A site taking payments that's a bit slow and out of date is a liability.

And some readers should skip all of this. If your distributors are your channel and you don't intend to compete with them, build the Where to Buy page and stop there. If keying the occasional card by phone works fine, a checkout is a fixed cost against volume you may not have.

Where we come in

We host and look after WordPress sites for small and mid-size companies, so we've watched a fair number of these projects go both ways — including the ones where the honest answer was a single page and a phone.

If you're at the four-ideas-and-no-first-move stage, that conversation is free. Tell us what you sell and who sells it for you, and we'll tell you which job is worth doing first.