Published: November 21, 2025
Historically, browser support on Target.com has been based on primarily supporting all users that make purchases on Target.com. This policy changes at major decision points, like stopping support for Internet Explorer entirely, or targeting a certain minimum version of a browser to gain access to a high value feature. This happens once every couple of years when the need arises.
Without having a progressive policy on what browsers and features to target, Target.com only allowed modern features in the codebase using heavy-handed solutions like polyfilling and transpiling to very dated versions of JavaScript. When the WebDX Community Group launched Baseline, stakeholders in Target.com saw it as the right time to think about finding a more appropriate minimum support target.
With Baseline, Target now confidently knows what features are available in supported browsers, and can identify more recent features that have become available with progressive enhancement and polyfills as possible fallbacks.
The problem
Dozens of engineers contribute code to Target.com on any given day. It's common in code reviews to point out features that don't work with Target.com's supported browser versions using Can I use as a resource. When engineers continually receive comments to change their code to prefer legacy features rather than modern features, avoidance of new web features is the result. Target then falls back to using "old" techniques that work, but the opportunity to use modern features is delayed to another time. The usage of modern web features often delivers a better developer experience, and can also provide a better user experience by shipping less code.
A data-driven approach to browser support
Target.com has a webpack config that defines the minimum supported browser versions. It has been historically hard to justify bumping up these minimum supported browser versions. As of early 2025, the rules were:
- Current and previous two versions of Chrome, Edge, and Firefox.
- Safari 11 and greater.
Safari is treated with more care due to the high volume of traffic and sales Target gets from Safari on iOS. Initially, there was an educated decision to set Safari 11 as the minimum version to develop for. This decision pinned Target.com to web features that were available on and before 2017.
In the first step in Target's journey to integrate Baseline into their development workflow, a data-driven approach was used. Through research, Target found Safari versions 11 through 14 delivered very little business impact—specifically .0001% of demand sales on Target.com. Given this reality, Target recognized that removing transpilation and polyfills for these outdated browser versions would provide significant opportunities to improve site performance.
Additional research showed that Safari 15.4 was the first Safari version that delivered at least 0.5% of demand sales, and each minor version of Safari 15 after that had a similar impact. Anytime Target runs an A/B test, altering 0.5% of demand sales is very valuable, and leads to the conclusion that the minimum supported version of Safari should be somewhere within version 15.
An interesting trend we have found in this research is how quickly old Safari browser usage is moving. As of September 2024, Safari 15 contributed only to 0.94% of demand sales on Target.com. In January 2025, it was 0.67% of demand sales, by May 2025 it had diminished further to 0.45%, and by November 2025 it was at 0.32%. What Target learned from this is that, if a real money threshold as a percent of demand sales site-wide is determined, support for these browsers can end in an automated way, and the next major version on Safari 16 can be achieved by the end of the year.
Dropping support doesn't mean that unsupported browsers are blocked—people who are using them may still be able to find a path to purchase, but they may also encounter a reduced experience in some cases. Continuing to use the data-driven approach, after changes were made, analysts reported no measurable impact to business metrics. Target is additionally looking into a banner that displays on unsupported browser versions warning of the degraded experience.
Choosing a Baseline target for Target.com
Target's web engineers formed a Baseline working group to combine these efforts to use modern features and the moving target of which browsers to support to help drive the policy. Using Baseline's tooling, they looked at what each year's minimum browser set was. The closest target to Target's new policy was Baseline 2022:
{
"chrome": "108",
"edge": "108",
"firefox": "108",
"ios": "16",
"safari": "16"
}
To get to that Baseline, Target would need to adjust its browser policy to Safari 16 at a minimum instead of the current version of 15.4. This would degrade the experience of less than 0.5% of converting shoppers. Even so, that percentage is shrinking, so Target hopes to update its official policy for now to be tied to Baseline 2022 by the end of 2025. This puts developers for Target.com in a position where the target can be changed to be around 3 years behind the annually published Baseline.
Overall, the webpack bundles for Target.com are smaller due to less transpiling of code and added polyfills. Target is confident that this target will move over time, and hopefully, at this time next year, Baseline 2023 can be adopted, which includes many great features like container queries, the :has selector, the inert attribute, and more.
Looking at newer Baseline feature sets
The Target Baseline working group isn't stopping with Baseline 2022. Looking at the features of Baseline 2023, a lot of them are just on the threshold of being able to support without backups like polyfills. Each one of the features in Baseline 2023 that Target is interested in requires them to:
- Articulate what the feature does.
- Document how its usage could improve Target.com, including developer experience improvements.
- Find a good test case to implement the feature within the Target.com codebase.
- If necessary, document what fallbacks to use including progressive enhancement, or other solutions provided through feature detection.
- Finally, when can usage of the feature be expected to be approved? Can it be used now? Or should some future threshold be waited for?
One example of this is the inert attribute. The minimum version to use inert in Safari is 15.5, meaning that Target.com is close to being able to use it! Target.com has many modal implementations where this attribute would be a benefit compared to its current JavaScript solution. Having an engineer write up the report on this feature allows for knowledge sharing and preparation for the next relaxing of the browser policy. This helps to make the case that dropping support for a browser version that brings little business value can unlock features that do have value. The feature can be engineered, reviewed, and deployed under a feature flag, and be ready in the event that it can be used.
In addition to this, a different engineer will follow the same process for using container queries—which is now Baseline widely available. Container queries could be used with a polyfill, but the polyfill has known performance issues. The solution Target came up with was to use container queries only as a progressive enhancement until browser minimums rise to fully support the feature.
This process works well for Target.com, because when the point arrives that the feature's minimum version is in sufficient use, progressive enhancement is no longer needed, and the feature can be used. During a recent audit, it was found that Target.com was shipping so many unnecessary polyfills that implementing Baseline into their application can help to keep this sort of technical debt in check.
Correlating Baseline concepts to web performance
Performance is important to any retail website. One belief shared by developers working on Target.com is that too much JavaScript is shipped. If 5% of the JavaScript bundles shipped to users was eliminated, it's a big win—but it also doesn't significantly improve Core Web Vitals across Target.com. Though if Target were to accomplish this 10 times, a 50% decrease in bundle sizes would be realized, and that would significantly contribute to Target's performance goals.
When it comes to Target's approach to Baseline, it has allowed engineers on Target.com to start thinking about the amount of JavaScript that's relied on for things like modals, accessibility needs, popovers, carousels, accordions, and other common user experience concerns. Each of these require polyfills or custom JavaScript solutions that contribute to an application's JavaScript bloat. As Target uses Baseline, browser targets evolve over time, and policies to include newer features can be relaxed. Target expects to transpile less code over time, polyfill less features, and even adopt web components when the opportunities arise. By paying attention to polyfills and targeted browsers shipped in project toolchains, the JavaScript bundle size of Target.com has already been reduced by 10%. This is before adopting any newer features. This should improve year over year, and this directly correlates to big bets Target is making in performance improvements for Target.com.
Takeaways
Having a Baseline target and very well-curated Baseline reports of Newly available and Widely available web features has been a powerful tool for Target.com. Here are some key results:
- Browser target has moved from supporting browsers released 8 years ago to 3 years ago.
- Baseline target of Baseline 2022 will be reached by the end of 2025.
- JavaScript bundles total size of Target.com has been reduced by 10%.
- The long tail of old browsers providing less than 1% of business is reducing at a rate of approximately 300% per year (0.94% in September 2024 to 0.32% in November 2025).
The realization that the web is moving faster than ever has energized Target to adopt features as fast as ever. Organizing these features allows for advance work and planning for when Target unlocks each one, and gives them confidence to trust their engineers contributing to a large retail website to know which features to use and when they can use them.