Conversion rate optimization (CRO) is the practice of getting more of your existing visitors to buy, by finding out why they don't and testing changes that fix it. For an online store, that means researching shoppers, forming a hypothesis, running an A/B test on a page like the product page or cart, and keeping only the changes that measurably raise conversion rate or revenue per visitor.
This guide covers how that works in practice, with real results from the LiftArchive test database so you can see what CRO looks like when people actually measure it.
What is conversion rate optimization? CRO in one paragraph
Your conversion rate is the share of sessions that end in an order. CRO is the work of raising that number, or the closely related revenue per visitor (RPV), without buying more traffic. It is not a list of tricks. It is a loop: you study where and why shoppers drop off, change one thing, measure it against the original, and learn from the result.
A useful reference point: Littledata's benchmark shows a median conversion rate of 1.4% across 436 Shopify stores for July 27 to October 24, 2026, with 2.7% or higher putting a store in the top 20%. See our Shopify conversion rate benchmarks for more detail.
The gap between 1.4% and 2.7% is the business case for CRO. Doubling conversion on the same ad spend roughly doubles revenue from that traffic. New to the terms? Our CRO glossary defines CVR, RPV, AOV and the rest.
Why revenue per visitor matters more than conversion rate
Many practitioners in our database judge tests on RPV, not just conversion rate. A change can lift orders and still lose money if it pushes people toward cheaper baskets.
Dave Diederen's pre-checked subscription test shows why. It produced -9.17% RPV upfront but a +1,295% increase in subscription share. Whether that is a win depends on lifetime value, and that judgment is part of CRO too.
The CRO process: research, hypothesis, test, learn
Every serious CRO program follows the same four steps. Our full CRO process guide goes deeper, but here is the short version.
1. Research
Start with evidence, not opinions. The main sources are analytics (where people drop off), heatmaps and session recordings (what they do on the page), customer surveys and reviews (what they worry about), and your own past tests.
Research is where good tests come from. Dave Diederen changed the above-the-fold USPs to the points heatmaps showed buyers engaging with and got +6.23% CVR. Oliver Kenyon put three survey-based reassurance lines above a collection page grid, a test logged at $44,890/month.
You don't need a big budget for this step. Microsoft Clarity is free, offers heatmaps and session recordings, and states "no limits on traffic."
2. Hypothesis
Turn each finding into a testable statement: "Because we saw X, we believe changing Y will cause Z, measured by metric M." A clear hypothesis tells you what to build and how to read the result.
Then rank your ideas. Effort, expected impact and how much traffic the page gets all matter. A simple score for each idea is enough to decide what goes first.
3. Test
Split traffic between the original (control) and your change (variant), and run until you have enough data to trust the result. The basics are in our A/B testing guide, and the sample size calculator tells you how many visitors you need before you start.
Stopping early is the most common way to fool yourself. Ruben tested an expert endorsement badge that looked like +17% after week one. After four weeks it was flat at +1.66% CVR, and the early spike was noise.
4. Learn
Write down every result, including losses and flat tests. A loss tells you something about your shoppers, and it stops you from repeating the same mistake.
Countdown timers are the clearest example of why context matters. Dave Diederen saw +3.16% RPV (+$139,029/month est.) from a countdown timer in checkout, yet another checkout timer test lost at -7.1% RPV. A timer for a European market lost -35.61% RPV. Kasper Dolk removed a checkout countdown timer and got CVR +25.61% and RPV +26.18%, est. +€150k/month.
The same idea won and lost across stores. That is why CRO depends on testing, not copying.
What gets tested in ecommerce CRO
Most ecommerce tests happen on a handful of page types. Here is what our database shows for each, with links to the full hub pages. For a store-focused overview, see ecommerce CRO.
Product pages
Product pages get the most tests in our database by far: 128 of 285 results. That makes sense, since this is where shoppers decide.
Headlines are a frequent winner. Dave Diederen changed the product title to an "Achieve [outcome] with [product]" format and saw +15.61% CVR. Oliver Kenyon added 60-day guarantee and easy returns blocks under Add to Cart, a result logged at +€63,000/month.
Not every popular tactic works. Milos changed a 2-pack offer to "Buy 2 Get 1 Free" at the same price and lost -52.38% RPV (-$104,679/month). Browse all product page tests.
Homepage and navigation
The homepage often works best as a fast route to products. Dave Diederen added a best-sellers row right under the homepage hero and got +30.88% CVR, and replacing a hero video with a static image and benefit callout produced +52.67% RPV.
Navigation matters too. Oliver Kenyon's mega menu with category descriptions and a testimonial was logged at +$151,192/month.
Collection pages
Collection pages help shoppers narrow down choices. Dave Diederen let shoppers swipe through product images on collection cards for +10.49% CVR.
Cart and checkout
Small reassurance near the checkout button often pays off. Oliver Kenyon saw +23.8% revenue per visitor from a cart drawer with a dispatch, returns and payment block under checkout. Dave Diederen showed customer reviews on the checkout page and got +17.96% CVR.
Removing steps is not always better. Skipping the cart and sending shoppers straight to checkout lost -10.23% CVR.
Site-wide offers and shipping
Pricing and shipping rules affect every page. Dave Diederen tested shipping price levels and found a $9.49 fee beat the control at +43.81% RPV, and removing an email popup lost -20.51% RPV.
CRO vs SEO vs paid ads
CRO, SEO and paid ads solve different problems. SEO and paid ads bring people to your store. CRO decides how many of them buy once they arrive.
| CRO | SEO | Paid ads | |
|---|---|---|---|
| Main goal | More orders and revenue from current visitors | More free visitors from search | More paid visitors, fast |
| Main cost | Research time, testing tools, design and dev work | Content and technical work | Ad spend on every click |
| Time to results | Weeks per test | Months | Days |
| Does the gain last? | Yes, a winning change keeps working until you change it | Yes, while rankings hold | No, traffic stops when spend stops |
| Needs traffic first? | Yes, to test with confidence | No | No |
They work together. A better product page makes every ad click and organic visit worth more. That is why "CRO marketing" usually means using conversion work to get more out of the marketing budget you already have.
Message match ties the two together. Dave Diederen matched the social proof line under the add-to-cart button to the ad angle the visitor came from and saw +12.70% CVR. Oliver Kenyon tested new listicle landing pages vs the homepage for paid traffic and got +57% revenue.
Landing pages built for ads are a CRO tool in their own right. Dave Diederen moved a week-by-week results timeline higher up a listicle landing page and got +24.36% CVR. Oliver Kenyon found that a landing page rebuilt natively in Shopify beat the same page in a page builder by +9.9% conversion rate.
SEO and CRO can also pull in different directions. Long content helps rankings, but a product page stuffed with text can bury the buy button. Test changes made for SEO on your key pages the same way you would test any other change.
The trade-off: CRO needs enough traffic to measure. If you get very few visitors, more traffic may be the better first investment.
How to start CRO with a small store
Small stores can do CRO. They just can't run as many A/B tests, because each test needs enough orders to reach a reliable result. Here is a practical way to start.
Step 1: Set up measurement
Make sure you can see conversion rate, add-to-cart rate, checkout rate and RPV, split by device. Install a free heatmap and recording tool such as Microsoft Clarity. Then watch 20 or 30 sessions of people who left without buying.
Step 2: Fix obvious problems without testing
Broken buttons, slow pages, missing shipping info and confusing variant pickers don't need a test. Fix them. Save testing for real questions where the answer isn't obvious.
Step 3: Borrow ideas from tests other stores ran
This is where our database helps. Instead of guessing, start with changes that have won for other stores, then check them against your own research. A CRO audit is a good way to find which ones fit your store.
Some patterns have won often. Dave Diederen reports that showing the product price directly below the product title had a ~90% win rate across 20+ brands. Treat even that as a strong starting idea, not a guarantee.
Customer research matters even more when you can't test much. Email five recent buyers and five people who abandoned a cart, and ask what almost stopped them. Read your product reviews and support tickets for repeated questions. The same three or four worries usually come up, and they tell you what to fix first on the product page and in the cart.
Step 4: Run bigger tests
Low traffic means small changes are hard to detect. Test bolder changes, like a new product page layout, so the effect is large enough to measure. Milos rebuilt a best-seller product page and saw +128% RPV (+$14,867/month). Our guide to A/B testing with low traffic explains the options.
Step 5: Choose a testing tool
Shopify stores have a native option now. Shopify's Rollouts feature includes an Experiment type that compares a treatment against a control, and Shopify's help docs say your store must be on the Grow plan or higher to run experiments. See Shopify CRO for more.
Dedicated tools give you more control. As of November 2026, Convert's pricing page lists the Growth plan at $299/month billed annually ($399 month-to-month) for 100K tested users a month, with a 15-day free trial and no credit card required. That is a real cost for a small store, so Convert is not for stores still on low traffic.
VWO's pricing page lists Growth, Pro and Enterprise tiers for web testing but asks you to book a demo rather than showing prices. VWO bundles testing with heatmaps and recordings, which suits teams that want one platform. It is a poor fit if you want to compare prices quickly or only need simple theme tests. Our CRO tools directory compares more options.
Tools mentioned here
Hiring CRO help vs doing it yourself
You can learn CRO yourself, hire a freelancer or agency, or mix the two. The right choice depends on your traffic, budget and how much time you have.
Doing it yourself
DIY works when you have time, some design or dev skills, and a store with enough traffic to test. The cost is mostly your hours plus a testing tool. If you want structured training, see our list of CRO courses.
The risk is inexperience. Common errors include stopping tests early, testing tiny changes on low traffic and reading noise as a win.
Hiring a practitioner
A specialist brings a backlog of ideas that have already been tested elsewhere. That shortens the path to your first wins. Most of the results on LiftArchive come from practitioners who do this work for stores every week.
Dave Diederen, for example, has run many of the product page and checkout tests cited in this guide, such as the advertorial hero redesign that produced +12.26% CVR and $1M additional yearly revenue.
Want results like this on your store?
Dave Diederen at Syntra Labs. Dave builds and tests Shopify stores at Syntra Labs, and shares most results in public.
Oliver Kenyon focuses on revenue-level wins across the store, such as removing a live chat feature that was costing revenue (+$28,000/month).
Want results like this on your store?
Oliver Kenyon at Conversionwise. Oliver co-founded Conversionwise and has spent 13 years helping ecommerce brands turn clicks into customers.
How to decide
- Do it yourself if you are early, have low traffic, or want to understand your customers first-hand.
- Hire help if you have steady traffic, a dev bottleneck, or tests that keep coming back flat.
- Mix both by doing the research yourself and paying a specialist to design and run tests.
Before you hire, ask to see past test results with real numbers, including losses. Browse CRO experts on LiftArchive, or read how to choose a CRO agency.
Conversion rate optimization strategies and best practices
The tests above point to a few strategies that show up again and again in our database. None of them is guaranteed to work on your store, but they are good places to look first.
Lead with the outcome, not the product name
Benefit-led copy is one of the most consistent winners. Dave Diederen saw +10.47% CVR from rewriting a product title as a benefit-led headline. Kasper Dolk reports a benefit-led buy box proposition brought a +$141,673 revenue lift at 100% confidence.
Answer the worry at the moment of decision
Shoppers hesitate over shipping, returns and whether the product works. Putting the answer next to the buy button often helps. Dave Diederen saw +7.48% CVR from putting a "60-Day Money-Back Guarantee" inside the add-to-cart button, and +7.40% CVR from replacing a "Fast Shipping" badge with a "Ships by Tomorrow" line.
Use social proof where it is seen
Reviews work best near the decision. Dave Diederen logged +$47k/month from adding one written customer review above the fold. But more is not always better: adding small customer profile icons next to the star rating lost -3.84% RPV.
Remove what distracts
Many winning tests take things away. Dave Diederen hid an "As Seen In" press banner on the product page for +5.63% RPV. Oliver Kenyon removed a 1/2/3-bag bundle selector, a result logged at +€38k/month.
Test offers and pricing carefully
Offers can move revenue more than design, in either direction. Oliver Kenyon saw +48.71% revenue per visitor from adding a free BBQ recipe e-book with knife purchases. The Buy 2 Get 1 Free loss earlier in this guide shows the risk.
Habits that keep a CRO program honest
- Measure revenue, not only clicks. Track RPV alongside conversion rate.
- Test one clear idea at a time, or a whole redesign on purpose, so you know what caused the result.
- Run tests long enough. Decide the sample size up front and cover full weekly cycles.
- Segment by device and market. The European countdown timer loss above shows a winning idea can fail in another market.
- Keep a test log. Losses and flat results are useful data. Dave Diederen's test of a "Clinician Recommended" badge came back flat, and that is still worth knowing.
- Retest winners after big changes. A new theme, new market or new traffic source can change what works.
The fastest way to build intuition is to read what other stores tested. Start with the full test database and filter by the page you want to improve.
Questions
What is conversion rate optimization?
Conversion rate optimization (CRO) is the process of getting more of your existing visitors to buy. You research why shoppers drop off, form a hypothesis, A/B test a change, and keep only the changes that measurably raise conversion rate or revenue per visitor.
What are the best conversion rate optimization strategies for ecommerce?
Strategies that win often in our database include benefit-led product titles, answering shipping and returns worries next to the add-to-cart button, placing reviews near the buying decision, and removing distractions. Results vary by store, so test each idea rather than copying it.
What are some conversion rate optimization examples?
Dave Diederen saw +30.88% CVR from adding a best-sellers row under a homepage hero. Oliver Kenyon saw +23.8% revenue per visitor from adding a dispatch, returns and payment block under the cart drawer checkout button. Some ideas also lose: a Buy 2 Get 1 Free offer tested by Milos lost -52.38% RPV.
What is the difference between CRO and SEO?
SEO brings more visitors from search. CRO gets more of those visitors, and visitors from every other channel, to buy. They work best together, because a higher conversion rate makes every visit worth more.
Can a small store do CRO?
Yes. Start with measurement, fix obvious problems without testing, use customer research and tests other stores ran for ideas, and run bolder tests so the effect is large enough to measure on low traffic.