[Help] ~650 orders/mo shop — host says upgrade, dev says VPS, I still think Action Scheduler is the villain
Side business out of Austin TX, custom gift boxes on Woo, been at it about four years. Peak season (Oct–Dec) we float around 650–700 orders a month. Still on the same managed Woo plan from when I was doing 40 orders a month, ~$45/mo, shared CPU with a hard throttle when usage spikes.
Checkout randomly spins, admin order list takes forever, and Tools → Scheduled Actions shows hundreds of pending hooks after a busy weekend. PageSpeed on product pages still looks fine so my brain keeps blaming background junk instead of the storefront.
Already tried Cloudflare in front with cart/checkout bypass rules, DISABLE_WP_CRON plus a real cron every 10 min on the host panel, and clearing failed actions twice. Also disabled a social feed sync plugin that was enqueueing retries.
Host support keeps pointing at the plan tier. A dev friend keeps saying move to a small VPS so cron and MySQL arent fighting neighbors. I cant tell if Im paying for guilt or if shared is genuinely the ceiling at this order volume.
For stores in that awkward middle (not enterprise, not hobby): when checkout got flaky, did fixing the queue + real cron calm it down on shared, or did you only get relief after you left shared entirely? What was the thing that finally made the cause obvious in hindsight?