What to put in your online store, and what to leave out
The first thing most new store owners do is start adding things. Reviews, pop-ups, chat, upsells, loyalty points, a currency switcher, a wheel that spins for a discount. Each one looks like it might help, and each one takes a small bite out of your page speed, your monthly costs and your ability to work out what's actually driving sales.
A useful rule: nothing goes in the store unless you can say which number it moves. If you can't name the number, you're decorating.
Start with the numbers you're trying to move
There are only four, and everything you install should be attached to one of them.
Traffic. How many people arrive.
Conversion rate. What share of them buy.
Average order value. What they spend when they do.
Repeat rate. How many come back.
Multiply the first three and you have revenue for the period. Improve the fourth and you stop paying to acquire the same customer twice. Any tool that doesn't clearly serve one of these is costing you without a case for itself.
What earns its place early
Analytics and conversion tracking. Non-negotiable and first. Until you can see where people drop off, every other decision is guesswork. Install it, then verify it fires by making a test purchase and checking the event actually lands. A surprising number of stores run ads for weeks against tracking that was never working.
Email capture and a basic flow. Email is the only channel you own outright. Everything else rents you an audience. Two automated sequences do most of the work: a welcome message for new subscribers, and an abandoned-cart reminder for people who got as far as the checkout and stopped. Abandoned-cart email is usually the single highest-return thing a new store can add, because those people already told you they wanted the product.
Reviews. Social proof affects conversion more reliably than almost any design change. The catch for a new store is that you have none, and fake ones are both obvious and, in many places, illegal. Better to launch without and add them as real orders come in.
What can wait
Live chat. Useful once you have enough traffic to justify someone answering it. An unattended chat widget is worse than no chat widget, because it promises a response you won't give.
Loyalty and points programmes. These reward repeat purchase, which means they do nothing until you have repeat customers. Revisit at a few hundred orders.
Spin-to-win and aggressive pop-ups. They lift email signups and often depress the experience for the people most likely to buy. If you use one, at least keep it off the checkout.
Currency and language switchers. Only once you have real traffic from somewhere that needs them. Check your analytics before assuming you do.
Anything with an "AI" badge and a monthly fee. Ask what number it moves and how you'd know. Most of these answer neither question.
The cost nobody counts: speed
Every app adds code to your pages. Slower pages lose sales, and the effect compounds on mobile, where most of your traffic will be.
This is the hidden reason "just add it, it's only a few dollars a month" is bad advice. The subscription is the small cost. The load time is the big one, and it doesn't show up on any invoice.
Two habits help. Test your store on a real phone on mobile data rather than office wifi, and remove anything you installed and stopped using, because uninstalling from the dashboard doesn't always strip the code from your theme.
How to decide, in practice
Before adding anything, answer three questions.
- Which of the four numbers does this move?
- How will I know whether it worked, and by when?
- What does it cost me in monthly fees and in page weight?
If you can't answer the first two, don't install it. If the answer to the third is "I don't know", find out before you commit.
Then add one thing at a time and leave it running long enough to read the result. Installing five apps in a weekend means you'll never know which one did anything, and you'll keep paying for all five.
The order that tends to pay off
For a store starting from nothing, roughly this sequence:
- Analytics and conversion tracking, verified working
- Email capture with a welcome and an abandoned-cart flow
- Whatever your checkout needs to stop losing people, which you'll only know once step one is running
- Reviews, as soon as you have real ones
- Average-order-value tools such as bundles or shipping thresholds, once you have steady conversion
- Retention and loyalty, once you have customers worth retaining
Most stores skip to step five or six because those feel like growth. They aren't, not yet. Growth at the start is finding out where you're losing people and fixing that, and you can't do it without step one.