A logged-out visitor sees someone else's cart? A page cache keeps one copy per key. Seven documented routes put a signed-in page in a shared one. Search the anonymous HTML for wp-admin-bar. Above zero, it leaks. All seven, and the fix: xspeedcache.com/blog/cached-… #WooCommerce

Site slow only while wp-admin is open? WordPress polls your server every 15 to 60 seconds, for every open tab. With a few editors at work that is a lot of requests no visitor benefits from. Throttle it per context, up to 300. The panel: xspeedcache.com/docs/heartbe… #WordPress

Weight check green and mobile LCP still failing? Our own scan passes any page under 1.5 MB. Of the 375 sites here that passed, 78.7% still failed the 2.5s LCP threshold on the same report. The cliff is near 350 KB. The curve, 912 sites: xspeedcache.com/blog/page-we… #CoreWebVitals

Preloader said finished and your cache is still cold? On Plain permalinks xSpeed read the sitemap wrong and queued one page, not every page. Free 1.4.4 fixes it, resumes a stalled crawl, and can skip post types. How Crawl Now works: xspeedcache.com/docs/crawl-n… #xSpeedCache

Cleaned your WordPress database and nothing got faster? Expected. Our own docs call it housekeeping, not tuning: the real wins are page and object cache. It does stop the regrowth. Limit Post Revisions is off by default. Set it to 5. xspeedcache.com/docs/databas… #WordPress

Mobile score far below desktop? Across 1,110 WordPress pages graded on both, the grade differed on 61.6%, median 12.0 points. Under 12 is noise. The bad case is a narrow gap with both scores low. That one is your server. The triage: xspeedcache.com/blog/mobile-… #CoreWebVitals

Behind Cloudflare and your first byte barely moved? Of 556 WordPress sites we probed behind it, 330 return cf-cache-status: DYNAMIC. The edge forwards every request. Median TTFB: 386ms forwarding, 117.5ms answering. The probe: xspeedcache.com/blog/wordpre… #WebPerf

Page Cache switched on, and still nothing is served from cache? advanced-cache.php only runs when WP_CACHE is defined and truthy in wp-config.php. Miss either one and caching silently never happens. Health → Overview checks both live: xspeedcache.com/docs/cache-d… #WordPress

Homepage says HIT, every other page says BYPASS? A Plain permalink puts a question mark in every post URL, and the query-string gate refuses those first. The homepage has none, so it caches fine. Settings → Permalinks, pick anything else: xspeedcache.com/blog/only-ho… #WordPress

Converted every image to WebP and the page is still heavy? Across 368 WordPress homepages, 73.2% of delivered image bytes are wider than any phone can use. 26.0% are already WebP or AVIF and still too wide. Resize before you convert: xspeedcache.com/blog/webp-av… #WebPerf

Purged the cache and the page is still old? Four stores can answer one WordPress URL. A plugin's button reaches one. Fetch it twice, plain and with ?cachebust=. If only the second is new, something in front holds the old one. Name the layer: xspeedcache.com/blog/cache-p… #WordPress

Switched from gzip to brotli and nothing got faster? We fetched 159 WordPress documents twice and compared the wire bytes. On 50 of them (31.4%) brotli was bigger than the server's own gzip. Read the bytes, not the header. The method: xspeedcache.com/blog/wordpre… #WebPerf

Speed plugin bill moves when your traffic does? That follows from where the work runs. NitroPack optimises on its own servers, so it meters 4 things: sites, pageviews, CDN bandwidth, team members. WP Rocket meters 1. Both models: xspeedcache.com/blog/nitropa… #WordPress

Comparing two caching plugins by their hit rate? Across 2,030 of our scan reports, plugins that set a response header read as a hit 83.4% of the time. Plugins leaving an HTML comment: 20.7%. That gap is detection, not speed. How we measured: xspeedcache.com/blog/cache-h… #WebPerf

Server answers fast, page still paints late? In the newest 300 public xSpeed Scan reports: 63 paint later than 2.5s, 24 of them answered the first byte under 200ms. Caching buys the byte, not the paint. Images and fonts do. Scan yours: xspeedcache.com/scan/all/?ut… #CoreWebVitals

Picked your caching plugin by install count? The same plugin reads 7,000,000+, 285,305 or 313 depending on who did the counting. None of those answers for your site. Your own response header does. Three instruments, three winners: xspeedcache.com/blog/cache-p… #WebPerf

Tablet visitors getting your phone layout from cache? xSpeed Cache 1.4.3 adds an opt-in tablet cache: a third stored copy, and you pick whether tablets get the desktop or the phone page. It sits under Separate mobile cache. All of 1.4.3: xspeedcache.com/changelog/?u… #xSpeedCache

Search Console flagged your LCP right after you enabled caching? That figure covers the last 28 days, so a cache from Tuesday is a sliver of it. Subtract 28 days from the alert date and list what else shipped. Why it is rarely the cache: xspeedcache.com/blog/search-… #CoreWebVitals

Self-hosted video on your site? There is no lazy-loading for it, in any browser. Of 462 video tags we read across WordPress homepages, 243 declare no preload at all, so the browser decides how much to fetch. Set preload=none and a poster: xspeedcache.com/blog/lazy-lo… #WebPerf

Cloudflare Enterprise on your WordPress site, and you don’t need a Cloudflare account. With the xSpeed Cache add-on, your traffic goes through Cloudflare’s enterprise edge network. Setup takes about a minute. 👇 youtube.com/watch?v=N-Xb6PMO… #WordPress #Cloudflare #PageSpeed