Shopify growth services: how to work out which one you actually need

Shopify growth services are four distinct jobs sold under one label: conversion rate optimisation (more of your visitors buy), paid media (more visitors arrive), SEO and content (visitors arrive without a cost per click), and retention (each customer is worth more over time). Each of the four moves a different number, and three figures in your Shopify admin tell you which number is actually stuck on your store. This guide is the arithmetic and the free tools to work that out yourself, before you pay anyone for a growth service you may not need.

Which three numbers decide what to fix first?

Revenue on a Shopify store is sessions × conversion rate × average order value, so every growth service is an attempt to move one of those three terms, and the weakest term is the one worth paying to fix. In your Shopify admin, go to Analytics → Reports and open Conversion rate breakdown for the last 90 days; the overview dashboard above it gives you sessions and average order value for the same window. Write the three numbers down before you read any agency proposal, because the proposal will not tell you which one it is really aimed at.

The conversion funnel in that same report is more useful than the headline rate, because it localises the problem to a step. Shopify breaks the funnel into sessions → added to cart → reached checkout → converted, and the biggest percentage drop between two adjacent steps is your most expensive single problem. A store that gets 8% of sessions to add to cart has a product page problem; one that gets 30% to cart but loses two-thirds between checkout and purchase has a shipping cost, payment method or form problem instead.

One more number decides whether any of this is worth doing: contribution margin per order, which is average order value minus cost of goods, minus shipping you absorb, minus payment fees, minus the cost of returns. If that figure is thin, no growth service fixes it, and the real work is pricing, product mix or supplier terms. An agency that signs a growth retainer with a store whose margins are broken is selling a slower way to lose money, and it is the one thing none of the four services on this page can repair.

Which growth service should you buy first?

The whole answer is in one table, and the rest of this page is how to check each row against your own store. Read down the first column until a row describes your numbers, then stop.

Buy one service at a time and give each one a number to move, because bundled retainers make attribution impossible. If CRO, ads, SEO and email all start in the same month and revenue rises 20%, you have learned nothing about where next quarter's budget should go. Staggering starts by six to eight weeks costs a little momentum and buys you the ability to tell what worked.

Ask for the deliverable you keep rather than the report you receive, because reports stop arriving when the retainer ends. Code in your theme, documented sections, validated schema markup, ranking pages, a cleaned ad account with its history, and an email list you own all stay with you. Monthly slide decks do not, and an engagement whose main output is a deck has produced nothing durable.

What does CRO actually do for a Shopify store?

Conversion rate optimisation raises the share of your existing visitors who buy, and it is the only growth service that makes every other one cheaper at the same time. Taking a store from 1.2% to 1.6% makes every ad click, organic visit and email click worth a third more than it was before, across every channel at once. That is why CRO is normally the first service worth buying on a store that already has traffic, and the last one worth buying on a store that has almost none.

Install Microsoft Clarity before you pay anyone for conversion work, because it is genuinely free with no traffic cap, no recording limit and no paid tier, and it will find things no audit deck will. Clarity gives you session recordings, click and scroll heatmaps, and two signals worth opening it for on their own: rage clicks, where someone clicks the same element repeatedly because nothing happened, and dead clicks, where they click something that is not actually interactive. On a Shopify store those two reliably surface a variant swatch that does not respond, a size guide link that opens nothing, and a mobile filter that silently fails.

Filter Clarity's recordings to sessions that reached checkout and did not convert, then watch ten of them end to end — it takes about forty minutes and it is the highest-value unpaid hour available to a store owner. We have never audited a store where that exercise did not surface at least one thing nobody internally knew was broken, most often on iOS Safari, which is where Shopify theme JavaScript tends to fail first. Do this before buying a CRO retainer, because it tells you whether you need one and gives you a list to hand over if you do.

The Shopify CRO fixes that actually pay are unglamorous and mostly subtractive: a product page that answers sizing, materials, delivery time and returns in the page itself; a cart that updates without a page reload; a mobile price and add-to-cart that sit above the fold; fewer required checkout fields; the payment methods your market expects; and a search that returns results for the words in your customers' heads rather than the words in your product titles. Each of those is theme work, which means it is a one-off cost that keeps paying, unlike an app subscription that does the same job with third-party JavaScript. This is the half of CRO we do — send us the list once Clarity has given you one, and you get a fixed price per item rather than a monthly retainer.

Put the answers on the page before you reach for a chat widget, because static content is cheaper, indexable, and works for the visitor who will never open a chat bubble. A chat widget earns its place on a catalogue large enough that browsing genuinely fails — hundreds of SKUs with compatibility, sizing or specification rules a filter cannot express. If that is your store, the thing to build is an assistant grounded in your actual catalogue data so it cannot invent a product or a stock level, which is work we do alongside the theme side; a generic bot bolted on to a small catalogue is another script on the main thread and another thing to maintain.

A/B testing is where Shopify CRO budgets most often get wasted, and the reason is sample size, not effort. Detecting a move from 1.5% to 1.8% conversion at 95% confidence and 80% power needs roughly 28,000 sessions per variant — about 56,000 sessions in total before the test can tell you anything; a 1.5% to 2.25% move needs about 5,100 per variant. Put your own figures into Evan Miller's sample size calculator and check the result against your monthly sessions, because if the test needs more traffic than you get in a quarter, an agency proposing it is selling you a report, not a decision.

Below A/B-testable traffic, the honest method is fix-and-measure rather than test-and-learn, and it is perfectly defensible. Change one clearly broken thing, note the date, and compare 30 days before against 30 days after on the Shopify conversion rate report, accepting that seasonality makes it directional rather than proof. That is a weaker claim than a significant test, and it is the correct claim to make at that traffic level — any agency presenting a 60-session "winner" as statistically meaningful is either confused or counting on you being.

Can your margins afford Shopify ads?

Paid media buys traffic volume on demand, and speed is its one honest advantage: a Meta or Google campaign can put buyers on a product page this week, which nothing else on this list can do. That makes it the right tool for testing demand for a new product, clearing seasonal stock, or holding the top of the funnel while slower channels build. It is the wrong tool for fixing a store that cannot convert, because ads multiply whatever conversion rate you already have.

Work out your break-even acquisition cost before you talk to a media buyer, because it takes two minutes and it decides whether ads can work at all. Take average order value, subtract cost of goods, shipping you absorb, payment fees and the amortised cost of returns, and what is left is your contribution per order and therefore your maximum customer acquisition cost. On a $40 average order value with 60% gross margin and $6 of shipping and fees, contribution is $18, so $18 is the ceiling — spend more than that to get a first order and you are buying revenue at a loss.

Converting that break-even ceiling into a cost per click gives you the one number you can check against a live ad account, because CPC ceiling = acquisition cost ceiling × conversion rate. With an $18 ceiling and a 0.5% conversion rate, you can afford $0.09 a click — a figure no cold Meta or Google campaign will deliver, which means the store must convert better before ads are viable. At 2% conversion the same $18 ceiling allows $0.36 a click, which is at least in the range of real auction prices for some categories.

Break-even ROAS is just the inverse of your contribution margin, and knowing that makes most ROAS targets easy to judge. In the example above, contribution is $18 on a $40 order, a 45% margin, so break-even ROAS is 1 ÷ 0.45 = 2.2 — a 2.0 ROAS is a loss there, while on a 70% margin product break-even is 1.4 and a 2.0 is comfortably profitable. Any agency quoting you a ROAS target without first asking for your margins is quoting a number chosen to look good in a report.

Fix your measurement before you spend, or you will optimise against fiction. That means server-side tracking through Meta's Conversions API or Google's enhanced conversions rather than browser pixels alone, a consent banner configured so it does not silently block the tags you depend on, and a check that Shopify's own order count agrees with what the ad platforms claim. If the platforms report twice the orders Shopify did, the campaign decisions built on that data are guesses with a dashboard around them. We run Meta ads, and we set the measurement up ourselves before spending anything — server-side events firing once and correctly, GA4 and Search Console verified, platform numbers reconciled against Shopify orders — because an agency that buys media on data it did not check is optimising towards a number that is not real.

Being the developer and the media buyer on the same store removes the handover that usually costs the most money. When a campaign underperforms, the question is always whether the targeting is wrong or the landing page is losing people, and that question normally bounces between two agencies for a fortnight. We can read the Clarity recordings of the paid traffic, change the template, and have the fix live in the same week the data showed the problem — which is the version of paid media worth buying.

Judge a paid media agency on the first conversation, where three questions do the work: do they ask for your margins and returns rate, do they state an acquisition-cost ceiling rather than only a ROAS target, and do you keep the ad account and its history if you leave. Those are the questions we expect to be asked and the standard we hold ourselves to — the ad account is yours, opened in your name, and it stays with you with its full history if you stop working with us. A management fee is always small next to the spend it controls, so competence there is worth more than a cheaper retainer.

Which SEO problems are specific to Shopify?

SEO earns traffic you do not pay per click for, and its defining property is that it is slow to start and slow to stop — a collection page that begins ranking in month four is still ranking in month twenty, where paid traffic ends the day the card declines. That asymmetry puts SEO in a 12-month plan and keeps it out of a 60-day one. Treat anyone promising Shopify rankings in six weeks as selling something else.

Shopify generates several duplicate-URL patterns by default, and they are the technical issues worth checking first because they are specific, findable and fixable. Products are reachable at both /products/handle and /collections/collection/products/handle; variant URLs add ?variant= query strings; filtered collection URLs add /collections/handle/filter-value paths; and paginated collections repeat the same intro copy on every page. Crawl your own store with the free tier of Screaming Frog, which does 500 URLs at no cost, and check that each of those patterns resolves to a single canonical URL.

Structured data is the part of Shopify SEO with the most mechanical payoff, because it changes how your result looks rather than only where it sits. Run your product and article pages through Google's Rich Results Test and validator.schema.org, checking that Product markup carries a price and availability that match the page, that Review markup only appears where real reviews exist, and that FAQPage markup only appears on pages with real questions. Themes get this subtly wrong constantly — the two faults we find most are an offers block reporting the wrong currency and a price in the markup that disagrees with the one a customer sees. Both are Liquid fixes rather than app installs, and fixing schema and canonical tags is routine work for us if the validator gives you errors you would rather not read.

Three free accounts cover most of what a Shopify store needs to monitor organic search, and there is no good reason not to have all three. Google Search Console shows the queries you already rank for and the pages Google refuses to index; Bing Webmaster Tools matters more than its market share suggests because it feeds Copilot; and Ahrefs Webmaster Tools gives you a free backlink and site audit for a domain you can verify. The highest-return report of the three is Search Console's Queries tab filtered to positions 8 to 20, because those are pages Google already likes that need one round of improvement rather than a new page.

Content that earns Shopify rankings answers what buyers search rather than what the brand wants to announce, and the distinction is testable. Comparison pages, sizing and fit guides, materials and care, "X vs Y", and use-case pages all map to real queries with buying intent; company news and brand milestones map to none. Check any planned topic in Google Search Console for impressions you already get, or in Google's own autocomplete and "People also ask" boxes, before committing a page to it.

Answer engines now reward the same structure, and it changes how to write rather than what to write about. ChatGPT, Google's AI summaries and Perplexity quote passages that stand alone, so a paragraph that only makes sense after the three above it is a paragraph nothing will cite. Name the subject in each paragraph instead of opening with "this" or "it", put the direct answer in the first sentence and the reasoning after, and keep one idea per section — which is exactly how this page is written, and you can judge the technique by whether any single paragraph here makes sense pulled out on its own.

What is a slow store costing your other channels?

Store speed is not a growth service by itself, but it caps the return on every service you buy, which is why it belongs in the comparison. A mobile product page that takes six seconds to become interactive loses a share of every visitor your ads, email and rankings deliver — you pay the acquisition cost and lose the session anyway. Fixing it raises the yield of all four channels at once for a one-off cost rather than a monthly one.

Measure against field data from real visitors, not a lab test on your laptop, because the two disagree in the direction that flatters you. PageSpeed Insights shows both: the top section is Chrome User Experience Report data from actual Chrome users on your store over 28 days, and the bottom is a simulated Lighthouse run. The field section is the one that reflects what your customers on mid-range Android phones on mobile data actually get, and it is usually two to four times worse than the lab score your theme vendor quoted.

Shopify's own Online store speed report under Analytics → Reports is worth reading alongside that, with one caveat about what it measures. It runs Lighthouse against your home page and your highest-traffic collection and product pages and combines them into a score out of 100, plus a comparison against similar stores — so it is lab data, useful for tracking your own trend over time rather than for knowing what customers experience. Use it for the direction of travel and PageSpeed Insights field data for the absolute truth.

On Shopify the usual cause of a slow store is the app stack rather than the theme, which makes it diagnosable rather than mysterious. A reviews widget, a bundler, an upsell app, a chat widget and two analytics pixels each inject third-party JavaScript competing with your own for the main thread. Open the Coverage and Network panels in Chrome DevTools on a product page, sort by unused bytes, and attribute each third-party script to the app that loaded it — then check Online Store → Themes → Customize → App embeds for embeds still active from apps you have uninstalled, which is a common and completely free win. Where an app has to be replaced by code rather than just switched off, that is a scoped fix sprint — our Shopify offer sets out what app-bloat removal involves and what it costs.

Is retention cheaper than buying new traffic?

Retention raises what each existing customer is worth, and it is the cheapest of the four services because you are not paying to acquire the audience again. Find your repeat rate in Analytics → Reports → Customers — "first-time vs returning customer sales" is the report to open — and compare returning customer revenue against total revenue. A store where returning customers are under about 20% of revenue has more money in its existing list than in any new channel.

Klaviyo is the Shopify default for this and has a free tier worth knowing the exact shape of: 250 active profiles, 500 email sends a month, 150 SMS or MMS sends, Klaviyo branding on every email, and email support only for the first 60 days. That is enough to build and prove the three flows that produce most of the revenue — abandoned checkout, welcome, and post-purchase — on a small list before paying anything. Shopify's own built-in email tool is a weaker but genuinely free alternative if your list is small and your flows are simple.

Three automated flows produce most retention revenue on a typical Shopify store, and building them is a one-off job rather than a retainer. Abandoned checkout recovers people who already chose a product and stopped; a welcome sequence converts the subscribers your popup collects and otherwise wastes; and a post-purchase sequence asks for the review that feeds your Review structured data and times a reorder prompt to your product's actual consumption cycle. If none of those three exist on your store, that is the cheapest growth work available to you and no agency is needed to tell you so. We do not write lifecycle campaigns — that is copy and a Klaviyo account rather than development — but the store-side pieces around them are ours: a signup popup that does not wreck your mobile load time, and the Review markup that turns post-purchase reviews into stars on your search results.

Which of these do we actually do?

Five of the six rows in that table are work we do: CRO implementation, speed, technical SEO, and Meta ads with the measurement underneath them. What makes that combination unusual is not the list but the fact that one team owns both halves — most stores buy development from one agency and media from another, then spend a fortnight waiting for the two to agree on whose problem a bad week was. We build the store and we buy the traffic, so the diagnosis and the fix happen in the same place.

Four things turn up in almost every store we are sent, and they are the same four this page tells you how to find yourself: a theme fighting a dozen apps, a six-second mobile load, collections search cannot read, and nobody left to call because the agency that built it moved on. If your own diagnostic landed on any of those, the engagement is already defined — and it is a fixed price, not a retainer that renews until you notice.

We have been doing Shopify client work since 2022, and the builds are written up case by case with what changed and why. Every project is assigned one developer who scopes it and is still on it after launch. Work is quoted as a fixed price after a discovery call rather than as an open-ended retainer, and the written scope is yours whether or not you hire us. Where a large catalogue is itself the conversion problem, we also build retrieval chatbots grounded in your own product data, which is a separate service to the theme work.

The one thing we will not do is take an ad budget from a store the arithmetic says cannot support it. If your contribution margin and conversion rate put your affordable cost per click below what the auction charges in your category, running ads converts your budget into traffic and your traffic into nothing — and we would rather spend that month on the store and start the campaigns when they can actually pay. That is a conversation we have had more than once, and it is the reason the margin section sits above the ad section on this page rather than below it.

If you work through this page and conclude the problem is the store rather than the traffic, send us the URL with your sessions and conversion rate and we will tell you which fix moves the most money and what it costs. Discovery is a fixed-price written scope you keep whether or not you hire us, and it says plainly when we are the wrong fit. Everything above is runnable without us, and that is deliberate — the stores worth working with are the ones that already know their numbers.