01
CRO audit
Analytics, funnels, speed, accessibility and checkout reviewed against evidence, ending in a ranked list of fixes and test ideas.
FIG 01 · Conversion Rate Optimization
A CRO agency that ships the winning version, not just the report
Research, a CRO audit and A/B tests where your traffic allows, then our own engineers build the winning change into your site. Tracking, speed and accessibility sit in the same scope.
See how it works
FIG 02 · Quick answer
What does a CRO agency do?
A CRO agency improves the share of visitors who buy, book or sign up. Elgrith’s conversion rate optimization services start with a CRO audit of analytics, speed, accessibility and checkout, then run A/B tests where traffic allows, or ship evidence-backed fixes where it doesn’t, and build every winning change into your site.
- Research, testing and the build, in one scope
- 1 team
- Projects delivered since 2015
- 1,000+
- Clients across the USA, UK and India
- 250+
- Your code, data and IP
- 100%
FIG 03 · Why it matters
Most carts are abandoned. Most CRO playbooks are out of date.
From Baymard Institute, Google and web.dev, checked on 23 Oct 2026.
01
70.22%
Average documented cart abandonment
Baymard’s average across 50 studies. It also estimates that large ecommerce sites could see a 35.26% increase in conversion rate through better checkout design alone.
Source: Baymard Institute, cart abandonment rate (last updated 22 Sep 2025)(opens in a new tab)02
30 Sep 2023
Google Optimize was switched off
Google says Optimize is “no longer available as of September 30, 2023.” Any CRO plan still built around it needs a new testing tool.
Source: Google Optimize Help, sunset notice(opens in a new tab)03
12 Mar 2024
INP replaced FID as a Core Web Vital
Interaction to Next Paint now measures how fast a page responds to taps and clicks, which is exactly where slow checkouts and filters lose people.
Source: web.dev, INP becomes a Core Web Vital (updated 31 Jan 2024)(opens in a new tab)
FIG 04 · Test or fix
Do you have enough traffic to A/B test?
An A/B test needs enough conversions in each version to tell a real difference from noise. How many depends on your current conversion rate and the smallest lift worth detecting (the minimum detectable effect). We run that sample-size check before we promise any test.
| Your situation | What it means | What we do |
|---|---|---|
| Enough conversions to finish a test in an agreed time | A test can detect the lift you care about | A/B test, with the length and significance threshold agreed before launch |
| Too few conversions to finish a test | A test would run for months or end inconclusive | Research and fix: ship evidence-backed changes and compare before and after, with the caveats written down |
| Tracking is missing or doesn’t match orders | Any result would be measured wrong | Fix GA4 key events and the tracking plan first |
| Poor results on every device and audience | Often a speed, accessibility or broken-step problem, not a design opinion | Fix the defects directly; test the bigger ideas afterwards |
| People don’t want the offer at this price | A product or pricing problem | Tell you honestly; CRO can’t fix it |
We don’t publish a single traffic threshold, because the right one depends on your baseline rate and the lift you are looking for. The sample-size check is part of every CRO audit.
Read the ecommerce CRO checklist for 2026FIG 05 · What you get
Conversion rate optimization services
CRO services for ecommerce, SaaS and lead-generation teams: find what stops people, prove the fix, then build it properly.
02
A/B and multivariate testing
A/B, multivariate and split-URL tests, client-side or server-side with feature flags, each with a written hypothesis and a pre-agreed length.
03
Landing page optimization
Clearer offers, proof and forms on the pages your ads and emails send people to, tested or shipped based on your traffic.
04
Checkout optimization
Fewer fields, guest checkout, clear costs and accessible forms, reviewed against Baymard’s published checkout guidelines.
05
Ecommerce conversion rate optimization
Product pages, search, filters, cart and checkout on Shopify, WooCommerce, Magento or a custom store.
06
Lead-gen and SaaS sign-up flows
Demo requests, trials, onboarding and pricing pages, with each step tracked as a GA4 key event.
07
Personalization
Different content for new and returning visitors, segments or locations, only where the data shows it helps.
08
Tracking and analytics
A tracking plan, GA4 key events, consent mode and dashboards, so every test is measured against the same numbers.
Deliverables
11
deliverables, in writing
Included in every build, written into your fixed quote before work starts.
FIG 06 · The loop
What a CRO agency does, step by step
Conversion rate optimization is a loop: research why people drop out, write a hypothesis, test it (or fix it directly), build the winner, measure, repeat.
1. Research
- Read
- GA4 funnels, heatmaps, session recordings, surveys
- Find
- Where people drop out, and on which devices
2. Hypothesis
- Write
- “If we change X for Y, metric Z improves, because…”
- Rank
- ICE or PIE scoring in a shared backlog
3. Test or fix
- Test
- When the sample-size check passes
- Fix
- When traffic is too low or the defect is obvious
4. Build the winner
- Who
- Our engineers, in your real codebase
- Not
- A permanent script layer from the testing tool
5. Measure and repeat
- Report
- Wins, losses and inconclusive tests, with raw numbers
- Then
- The next hypothesis from the backlog
FIG 07 · CRO audit
What a CRO audit covers
Every CRO audit checks the measurement first, then the experience. A test built on broken tracking proves nothing.
06principles
GA4 key events and a tracking plan
Since 27 Mar 2024, Google calls the important actions in GA4 “key events”. We check they fire once, match your orders or leads, and respect consent.
Funnel drop-off
Funnel explorations by device, channel and new vs returning visitors, to find the step that leaks most.
Core Web Vitals
LCP, INP and CLS at the 75th percentile of real page loads, on the templates that make money.
Technical SEO and speedWCAG 2.2 checkout criteria
Redundant Entry (3.3.7), Accessible Authentication (3.3.8) and Target Size (2.5.8) all touch forms and checkout.
Website accessibilityCheckout guidelines
Cart and checkout reviewed against Baymard Institute’s published checkout research.
Heatmaps and session recordings
Consent-aware, with personal data masked, used to explain the numbers, not replace them.
FIG 08 · Honest answer
Is CRO worth it?
Sometimes not, and a good CRO agency should say so before taking your money. Here is when it isn’t worth paying for, and when it is.
06principles
Not worth it: too little traffic to test
If a test can’t finish in a sensible time, a testing retainer is wasted. A one-off audit and fixes are the better spend.
Not worth it: broken tracking
Fix measurement first. Otherwise every “win” is a guess with a chart attached.
Not worth it: a product or price problem
If people don’t want the offer, button colors won’t save it. We’ll tell you that.
Not worth it: nobody builds the winners
Tests that never reach the real site don’t pay back. That’s why our engineers build them.
Worth it: steady traffic and a clear funnel
Enough conversions to test, measurable steps, and a team ready to ship changes every sprint.
Worth it: obvious defects
Slow pages, broken mobile steps and inaccessible forms are fixable now, with or without a test.
FIG 09 · Method
How we test
We publish our method instead of a win rate. Every test follows the same rules, agreed with you before it starts.
06principles
Hypothesis backlog
Every idea written as a hypothesis with its evidence, scored with ICE or PIE, in a backlog you can see.
Sample size and test length agreed up front
We calculate the sample size and fix the test length before launch, so nobody stops a test early on a lucky day.
Significance threshold agreed
The statistical significance level is agreed in writing before the test, not chosen after the results are in.
QA on real devices
Every variant checked across browsers, devices and screen readers before traffic sees it.
Tools named, no partner claims
We work in the testing tool you already license (for example Optimizely, VWO, AB Tasty, Convert or GrowthBook) or help you pick one. We aren’t a partner or reseller of any.
Rollout
Winners are built into the codebase and released through your pipeline; losers are written up so the lesson isn’t lost.
FIG 10 · Build
We build the fix
Most CRO agencies hand over a slide deck. Our engineers ship the winning change in your real stack, with tests and a rollback.
07principles
Next.js and React sites
Changes shipped as reviewed pull requests, with performance checked before release.
Shopify, WooCommerce and Magento
Theme and checkout changes within each platform’s rules.
Ecommerce developmentShopify Plus stores
Checkout extensions and theme work for larger Shopify stores.
Shopify Plus developmentWhen the site needs more than fixes
A redesign planned from audit evidence, not taste.
Website redesign
FIG 11 · Checklist
How to choose a CRO agency
Ask every CRO agency on your shortlist these questions, including us.
06principles
Do they disclose their method?
Sample size, test length and significance threshold, in writing, before the first test.
Who builds the winners?
If the answer is “your developers”, budget for that work too.
Who owns the data and accounts?
Your GA4 property, your testing tool account, your results. You should keep them if you leave.
Do they promise a lift?
Nobody can guarantee a percentage before testing. Treat guaranteed lifts and unsourced win rates as a warning sign.
Do they report losing tests?
Most tests don’t win. A report with only winners isn’t the whole story.
FIG 12 · Measurement
Evidence, not vanity stats
We don’t quote a win rate or an average lift. This is what we report on every test, and what we won’t claim.
| What we report | What we won’t claim | |
|---|---|---|
| Primary metric | One metric per test, agreed before launch (for example purchases or demo requests) | A different metric picked after the results come in |
| Sample size | The calculated sample size and whether the test reached it | “Directional” wins from tests that stopped early |
| Significance | The threshold agreed up front and the result against it | Significance on a metric nobody agreed to |
| Losing and flat tests | Every result, with what we learned | Only the winners |
| Revenue impact | Measured change in your own analytics, with the caveats | Guaranteed lifts or ROI before testing |
Numbers in our reports come from your GA4 property and your testing tool, which you own.
FIG 13 · Process · 6 steps
Research it, prove it, ship it
A fixed written quote after a free call, an audit before any test, and winners built into your site by the same team. You own the data, the accounts and the code.
- Step 01
Free call
Tell us what you want to improve and what you have already tried. We ask about traffic, tools, team and deadlines first.
- Step 02
Review
Read-only access to GA4, your testing tool and your store or CMS. We check the tracking, run the sample-size check and spot the obvious defects.
- Step 03
Fixed written quote
A written scope and a fixed quote, with the timeline, tools, access and hand-over agreed before any work starts.
- Step 04
Sprints with weekly demos
Audit findings first, then tests and fixes from the ranked backlog, with each test’s status and each shipped change shown in a weekly demo.
- Step 05
Launch
Winning variants built into the codebase and released through your pipeline, with tracking re-checked after release.
- Step 06
Support
Ongoing testing from the backlog, or a hand-over of the backlog and tracking plan to your team, as agreed in your written scope.
FIG 14 · What you get
Everything included
Written into your fixed quote, so you know exactly what you are paying for before work starts.
11items
- Written scope and fixed quote
- CRO audit with a ranked list of fixes and test ideas
- GA4 key events and tracking plan, checked
- Sample-size check for your key funnels
- Hypothesis backlog with ICE or PIE scores
- Test plans with metric, length and threshold agreed up front
- QA of every variant on real devices
- Winning changes built into your codebase
- Core Web Vitals and WCAG 2.2 AA checks on changed templates
- Results write-ups, including losing tests
- Your data, accounts and code stay yours
FIG 15 · Tech & tools
The tools behind the work
Proven, well-supported tools, chosen for your project and set up in accounts your business owns.
- GA4
- Google Tag Manager
- Consent mode
- Looker Studio
- Optimizely
- VWO
- AB Tasty
- Convert
- GrowthBook
- Microsoft Clarity
- Hotjar
- PageSpeed Insights
- Shopify
- WooCommerce
- Magento
- Next.js
- WordPress
- Figma
FIG 16 · Who it’s for
CRO for teams that have the traffic and want it to pay.
Same method for every team: measure first, test where traffic allows, fix where it doesn’t, and build the winners.
3 audiences
Ecommerce
More of your visitors reach the order page.
Product pages, search, cart and checkout on Shopify, WooCommerce, Magento or custom stores, with the fixes built by our ecommerce engineers.
- Checkout and cart audit
- Product page tests
- Core Web Vitals on key templates
- Accessible checkout forms
SaaS, lead-gen
More trials, demos and qualified leads.
Pricing pages, sign-up and onboarding flows, demo forms and landing pages, each step tracked as a GA4 key event so tests measure what sales cares about.
- Sign-up and trial flows
- Demo and lead forms
- Landing page tests
- CRM-matched key events
UK teams
Conversion optimisation for UK brands.
UK projects run through our London client partner, with engineering at our New Delhi HQ. We write in UK English, plan testing around UK GDPR and PECR consent, and agree working hours in your written scope. Our London CRO agency page has the detail.
- London client partner
- UK English throughout
- Fixed written quote in GBP
- Consent-aware testing
FIG 17 · Working with us from the USA and UK
Built in New Delhi. Run on your hours.
Every project gets a point of contact in New York or London and an engineering team in New Delhi. Here is how today’s working days line up, live.
New York
North America
--:--ET
New Delhi team Your working day- Shared working time
- ≈ 3 hours today
- New Delhi is
- 10.5 h ahead
- Live calls, your time
- 9 a.m. to 12 p.m. ET
London
UK & Europe
--:--UK
New Delhi team Your working day- Shared working time
- ≈ 4.5 hours today
- New Delhi is
- 5.5 h ahead
- Live calls, your time
- 9 a.m. to 1:30 p.m. UK
Three ways to work together
- 01
Fixed-scope project
A written scope, one fixed price and a dated plan. Suits websites, apps and MVPs.
- 02
Dedicated team
Designers and engineers on your roadmap every month, billed monthly.
- 03
White label
For US and UK agencies: our work, your brand, unbranded deliverables.
Every engagement
- Fixed quote in USD, GBP or INR after a free 30-minute call
- NDA before discovery. You own all code, designs and content
- A working demo every sprint, not status slides
- Reply within one business day, calls inside your hours
Book a call
Calls on ET or UK hours
FIG 18 · Decide
Hire a CRO agency or build an in-house CRO team?
Both work. The right answer depends on how much you will test and whether you already have developers free to build the winners.
| In-house CRO team | CRO agency that builds (Elgrith) | Agency that only advises | |
|---|---|---|---|
| Best for | High-traffic sites testing all year | Teams that want research, tests and the build in one scope | Teams with free developer capacity |
| Who builds the winners | Your developers | Our engineers, in your codebase | Your developers |
| Ramp-up | Hiring an analyst, a designer and a developer | Starts with the audit | Starts with the audit |
| Data and accounts | Yours | Yours | Check the contract |
| Works alongside the other | Yes: we can support an in-house team | Yes | Yes |
FIG 19 · Compare
CRO audit, testing program or redesign: which do you need?
| CRO audit | Testing program | Website redesign | |
|---|---|---|---|
| What it is | A one-off review with ranked fixes and test ideas | An ongoing loop of tests and builds from the backlog | A rebuilt site, planned from audit evidence |
| Traffic needed | Any | Enough to pass the sample-size check | Any |
| Best when | You don’t know why people drop out | You have traffic and a steady stream of ideas | The platform or design blocks every fix |
| What you get | Audit report, backlog, tracking fixes | Tested, shipped changes and write-ups | A new site with tracking and speed built in |
| Our page | You are here | You are here | Website redesign |
FAQ
Frequently asked questions
Straight answers about Conversion Rate Optimization with Elgrith.
A CRO agency increases the share of your visitors who buy, book or sign up. It researches where people drop out (analytics, recordings, surveys), writes hypotheses, runs A/B tests where traffic allows and ships fixes where it doesn’t. At Elgrith the same team also builds the winning changes into your site.
FIG 20 · Related
Keep exploring
Free 30-minute call
Find the leak. Ship the fix.
Tell us where people drop out, or where you think they do. After a free call and a look at your analytics, you get a written scope, a test-or-fix plan and a fixed quote.
Pick a day for your call
OpenPick a day. You'll choose a time in your own time zone on the next step.