Last Updated: 08/08/2026
Google Website Optimizer was Google’s original free A/B testing tool, launched in 2006 to help advertisers test different versions of a landing page.
It was renamed Google Optimize in 2012, grew into an enterprise tier called Optimize 360, and was shut down completely on 30 September 2023. If you are reading this because you found an old bookmark or a login screen that no longer works, that is why. The good news is that testing your website is far from dead. There are strong paid and free alternatives, and Google itself has quietly started rebuilding a tool with the same name inside Google Ads. We will cover both below.
A Short History, From 2006 To Today
Google first unveiled Website Optimizer at the eMetrics summit in October 2006, pitching it as a way for AdWords advertisers to test landing page layouts and see which version earned more conversions. It was basic by today’s standards. You tagged sections of a page, Google split your traffic between variations, and you read the results in a fairly clunky report. Even so, it was free, and free testing tools were rare at the time, so it caught on quickly with small businesses running their own PPC marketing agency campaigns and agencies alike.
In 2012, Google rebuilt the product from the ground up and renamed it Google Optimize. It gained a proper visual editor, tighter integration with Google Analytics, and a paid version, Optimize 360, for larger accounts. For the best part of a decade Optimize was the default starting point for anyone learning A/B testing, and it fed neatly into wider work on why the digital marketing industry keeps growing as more businesses moved budget online. Then, in January 2023, Google announced it was pulling the plug. Both the free and paid versions stopped working on 30 September 2023, and any experiments still live on that date were switched off for good.
What Google Website Optimizer Actually Did
It is worth understanding what the tool offered, partly because the concepts have not changed even though the software has. Website Optimizer, and later Optimize, supported three main types of test.
- A/B testing. You show one half of your visitors the original page and the other half a variation, then measure which one converts better.
- Multivariate testing. You test combinations of several elements at once, such as headline, image and button colour, to see which combination performs best.
- Split URL testing. You compare two completely separate page designs living on different URLs rather than tweaking one page.
Every result was measured against a goal, usually a sale, a sign up or a form submission, and the reporting sat inside Google Analytics once the two accounts were linked. If you have read older guides to optimising landing pages for better conversion rates, this is the toolset they were usually describing.
Why Google Shut It Down
Google’s official line was that Optimize “does not have many of the features and services that our customers request and need for experimentation testing”, and that the company preferred to invest in third party A/B testing integrations for Google Analytics 4 instead. Industry commentary at the time was more blunt. Optimize had not been meaningfully developed in years, it capped users at five concurrent experiments, and it had no heatmaps or session recordings, which put it well behind dedicated CRO platforms. The shutdown was widely reported by trusted industry outlets such as Search Engine Roundtable, which confirmed the exact sunset date and Google’s own FAQ on the matter.
For anyone running paid traffic, this left a real gap. Testing ad copy is only half the job. If you cannot also test what happens after the click, you are guessing at the part of the funnel that usually has the biggest impact on cost per acquisition. We see this constantly when reviewing accounts for clients, including when we walk through how to measure Google Ads performance properly, or when we look at creative strategies for increasing ad click through rates with a client.
The Name Is Back, But It Is Not The Same Tool
Here is the twist, and the reason this topic is worth revisiting in 2026. Late in 2025, new Google Ads help center documentation started appearing for a tool called Website Optimizer, reusing the original 2006 name. According to reporting from Search Engine Land, the tool is expected to sit under the Reporting tab in Google Ads, will require admin access to a linked Google Analytics 4 property, and will automatically create a GA4 property if one does not exist. Early documentation suggests it leans towards automated page suggestions tied directly to ad campaigns, rather than the manual, build your own variant approach of the original.
As of writing, it is still rolling out gradually and is not visible in every account, so treat it as a feature to watch rather than something to plan a whole strategy around just yet. If it does mature into a proper testing suite, it will likely sit alongside your existing Google Ads agency London setup rather than replace independent CRO tools entirely, since Google has historically kept its testing features tied fairly closely to its own ad platform. The same logic applies whatever you are selling, including examples we cover in converting website visitors into paying customers with incentives.
Best Google Optimize Alternatives You Can Use Right Now
If you need to test your website today, you do not have to wait and see what Google ships. The table below covers the main options, with prices converted to pounds and rounded, since most providers still bill in US dollars.
| Tool | Best for | Roughly from | Worth knowing |
| VWO | Marketers who want a point and click editor | About £145 a month | Closest experience to the old Optimize visual editor |
| Optimizely | Large organisations running many tests at once | Custom pricing, sales led | Strong statistics engine, built by Google’s original AdSense team |
| PostHog | Product and engineering teams | Free tier, then usage based | Combines testing with analytics, feature flags and session recordings |
| Crazy Egg | Small businesses wanting heatmaps plus simple tests | About £20 a month | Good entry point if your budget is tight |
| Convert | Ecommerce sites with privacy requirements | Custom pricing | Built with cookie consent and GDPR in mind from the start |
| Google Analytics 4, manual split testing | Solo marketers and very small budgets | Free | No built in experimentation, but workable if you are comfortable doing the maths yourself |
Prices are approximate, converted from US dollars and correct at the time of writing. Always check the provider’s own pricing page before committing, since most of these platforms adjust tiers fairly often.
How We Approach Testing Without A Free Google Tool
Losing a free tool is annoying, but it rarely changes the fundamentals. When we run tests for clients, whether that is refining a page linked to a Facebook ads agency London campaign or a B2B landing page built for LinkedIn, the process looks the same regardless of which software sits underneath it. Pick one variable, write down what success looks like before you launch, run the test long enough to reach a sensible sample size, and resist the urge to call a winner after three days because the graph looks exciting. We have written more on this in our piece on why LinkedIn ads sometimes fail to convert, and the same discipline applies whether the traffic source is search, social or email.
One thing that genuinely changed for the better after Optimize disappeared is that most modern platforms now record session behaviour alongside the test itself, so you can actually watch where someone hesitated rather than just seeing a conversion number move. That single addition has probably done more for our clients’ ad targeting and conversion rates than any statistics engine upgrade.
Why Testing Still Matters For UK Websites
The UK ecommerce conversion rate averages around 3.4%, comfortably ahead of many other markets, according to benchmarking data compiled by VWO. That sounds healthy until you remember it also means the average UK online shop still loses roughly 96 or 97 out of every 100 visitors without a sale. Even a modest lift, say from 3.4% to 3.8%, can translate into a meaningful revenue increase on the same ad spend, which is exactly why successful ad campaigns and conversion strategies almost always include an ongoing testing plan rather than a one off redesign.
If your site currently sits below that benchmark, the fix is rarely a complete rebuild. It is usually a handful of smaller, testable changes such as clearer calls to action, faster load times, or better alignment between the ad someone clicked and the page they land on. We cover the technical side of this in our guide to technical SEO for website health and performance, since a slow page will quietly sabotage even the best tested design.
Getting Started With Website Testing
You do not need Google’s tool, old or new, to start improving your conversion rate this month. A few practical steps that consistently work for our clients:
- Pull your Google Search Console and Analytics data first, so you know which pages get real traffic before you spend time testing low volume ones. Our guide to using Search Console for effective SEO covers the basics if you have not touched it before.
- Check page speed before you test anything else, since a slow page will suppress conversions regardless of what the copy says. Our piece on how to get on page one of Google touches on why speed and rankings tend to move together.
- Match your landing page message to the ad or search result that brought the visitor there. Mismatched messaging is one of the most common conversion killers we see, and we go into more detail on this in our LinkedIn focused guide to optimising landing pages for ad campaigns.
- Keep your site secure and stable while you test, since a hacked or slow WordPress install will skew your results before you even get to the marketing questions. Worth a read if you manage your own site: keeping a WordPress website safe from malware and hacking.
- Borrow ideas from real customer language rather than guessing. Platforms like Reddit are surprisingly useful here, and we explain how in our guide to using Reddit for competitive PPC analysis.
- Treat testing as ongoing, not a project with an end date. Search behaviour keeps shifting, and our notes on adapting SEO strategies for voice and AI search are a good example of why the goalposts move faster than most teams expect.
If you sell online, it is also worth reading our sector specific advice on SEO tips for ecommerce websites and our broader look at where to focus your SEO strategy, since testing works best when it sits inside a wider, well structured plan rather than as an isolated tactic.
Frequently Asked Questions
Is Google Website Optimizer still available?
No. Google Website Optimizer became Google Optimize in 2012, and both Google Optimize and Optimize 360 were shut down permanently on 30 September 2023. Any historic data not exported before that date is no longer accessible.
What replaced Google Optimize?
Google never built A/B testing directly into Google Analytics 4, despite hinting it would. Instead, businesses moved to third party platforms such as VWO, Optimizely, PostHog and Convert. Separately, Google Ads has begun testing a new tool that reuses the Website Optimizer name, though it is a different, more automated product tied to GA4 and still in gradual rollout as of writing.
Was Google Website Optimizer free?
Yes, the original tool and the standard version of Google Optimize were both free. Optimize 360, the enterprise tier with higher traffic limits and priority support, was a paid product typically bundled with Google Analytics 360.
Can I still A/B test my website without a Google tool?
Yes. Every alternative listed in this guide can run A/B, multivariate and split URL tests. Some, like PostHog, offer a genuinely usable free tier, so cost is not the barrier it might first appear to be.
Where This Leaves You
Google Website Optimizer had a good run, and the concepts it popularised, structured testing, clear goals, and data over opinion, are more relevant now than they were in 2006. The tool itself is gone, its 2012 successor is gone too, and whatever Google Ads eventually ships under the same name will likely serve a narrower purpose than a full CRO platform.
If you would rather not manage testing tools, traffic sources and landing pages on your own, get in touch with our team about our SEO services or our LinkedIn advertising London campaigns, and we will help you build a testing plan that does not rely on Google keeping a free tool alive.
