How do I know if my Shopify performance problem is severe enough to justify hiring an agency versus something my in-house team can fix in a day?
Short answer
A Shopify performance problem needs an agency only when the fix needs profiling or code you cannot edit, like theme JavaScript. In Deploi's judgment, leftover app code, oversized hero images and dead tag-manager tags are a one-day in-house job. App code lingers because uninstalling an app "doesn't automatically remove its code from your theme" (Shopify Help Center, October 2026).
- Good at the 75th percentile
- LCP 2.5 s, INP 200 ms, CLS 0.1Largest Contentful Paint, Interaction to Next Paint and Cumulative Layout Shift; Shopify's reports rank your store Good, Moderate or Poor on each from real visitors.Sources: Shopify Help Center (Web performance reports); web.dev (Web Vitals)
- Shopify's field data window
- Past 30 days, up to 36 hours lateShopify's metric summary ranks real-user data from the past 30 days by default, at the 75th percentile; the reports keep 90 days and annotate app installs, theme updates and new code.Source: Shopify Help Center (Web performance reports)
- Usual cause of poor INP
- Too much JavaScriptShopify traces poor Interaction to Next Paint to theme code, app code or third-party and tag-manager code, and says to clean up the worst offenders.Source: Shopify Help Center (Improving your online store performance)
- A speed app's monthly price
- Hyperspeed, $59 a month4.8 stars on 174 reviews on the Shopify App Store (11 October 2026). It defers, preloads and lazy-loads; Deploi's site speed record says lab scores rise while field metrics stay flat.Sources: Rvere (Hyperspeed EXTREME Page Speed on the Shopify App Store); Deploi Build vs. Buy (Build or Buy Site Speed Optimization on Shopify?)
Deploi’s call
Severity comes down to attribution: a problem whose file or app you can name is in-house work, and one that takes profiling to name is agency work. A fix is usually small once you know its source, so the first hour of an agency engagement should go to attribution, not optimization.
WaitHire nobody yet while
- Field data sits inside Good on all three Core Web Vitals and is not trending down, where Deploi's performance monitoring record says to wait and re-check quarterly (October 2026).
- The complaint started with a lab score or a competitor's screenshot, not Shopify's field reports.
- The four in-house fixes are not done yet: leftover app code, the tag manager, the hero image and section count.
BuildBring in an agency when
- Interaction to Next Paint (INP) fails, which Shopify traces to "too much JavaScript" across theme, app and tag code (Shopify Help Center, October 2026).
- The fix sits in a third-party theme, where Shopify sends you to the theme developer or a Shopify Partner (Shopify Help Center, October 2026).
- Vitals fail or trend down on a stack past roughly ten apps or two years without a cleanup, where Deploi's monitoring record scores an audit BUILD (October 2026).
- Regressions return after two fixes, because app-injected scripts "land in the theme without a deploy" (Deploi's script governance record, October 2026).
Basis: Build or Buy Performance Monitoring & App Audits on Shopify?, our Build vs. Buy decision record, scored BUILD.
Which data shows how severe a Shopify performance problem is?
Field data shows how severe a Shopify performance problem is, because it measures real visitors rather than a test run. Shopify's web performance reports show Largest Contentful Paint (LCP), Interaction to Next Paint (INP) and Cumulative Layout Shift (CLS) at the 75th percentile by default (Shopify Help Center, October 2026). Shopify's metric summary ranks real-user data from the past 30 days by default, the reports keep 90 days, and the data "might be delayed by up to 36 hours." Google's Good thresholds are LCP within 2.5 seconds, INP of 200 milliseconds or less and CLS of 0.1 or less (web.dev, October 2026).
A synthetic score from a testing tool is a diagnostic, not a verdict, and chasing one is how teams pay for gains no shopper notices. Whether Plus is a speed fix has a short answer, no, so the same work applies on every plan.
Which symptoms point to an in-house fix, and which to an agency?
Symptoms in Largest Contentful Paint (LCP), Interaction to Next Paint (INP) and Cumulative Layout Shift (CLS) point to who should fix them. Images and leftover code are in-house work, because fixing them needs no profiling; theme JavaScript and widgets you cannot edit need an agency. Shopify says poor INP most likely means "too much JavaScript" from theme, app or tag-manager code (Shopify Help Center, October 2026).
| Symptom in field data | Likely cause | Who fixes it | Realistic time |
|---|---|---|---|
| Slow LCP on product and collection pages | Images served far larger than shown | In-house | Under a day |
| Slow LCP on every template | Render-blocking third-party code in the page head | In-house; agency if an app injects it | A day to a sprint |
| Poor INP, LCP fine | Theme JavaScript or a heavy widget | Agency | Weeks |
| Poor CLS on collection pages | Missing image sizes, late banners and badges | In-house | Under a day |
| Regression with no deploy | An app's embed update or a new tag-manager tag | In-house to find; agency if an app | Hours to find |
| Slow on one template only | That template's custom sections | Agency | Weeks |
| Field data good, lab score bad | Nothing | Nobody | None |
What should an in-house team fix before calling anyone?
An in-house team should run four fixes before calling anyone, because each takes a day or less and needs no outside profiling.
- Inventory leftover app code. Compare theme app embeds and theme snippets against installed apps, because uninstalling an app "doesn't automatically remove its code from your theme" (Shopify Help Center, October 2026).
- Audit the tag manager. Shopify's advice is to "Audit your tag manager to remove any unused or low-value tags." Marketing usually knows which campaigns ended.
- Fix the hero image. Serve it at its displayed size and never lazy-load it. web.dev warns that lazy-loading the Largest Contentful Paint (LCP) image "will always lead to unnecessary resource load delay" (web.dev, October 2026).
- Count sections. Shopify warns that "High numbers of sections on pages can lower web performance." Homepages collect sections the way garages collect boxes.
Then wait for the field data to refresh. Shopify's metric summary ranks real-user data from the past 30 days by default (Shopify Help Center, October 2026), so a fix takes about a month to show fully; tell stakeholders before they check on Friday.
What thresholds justify hiring an agency?
Four thresholds justify hiring an agency, and each one means the fix needs profiling or code your team cannot touch.
- The fix is in code you did not write and cannot edit. A review widget or chat launcher that blocks rendering can often only be replaced, and replacement brings merchandising and legal inputs. Shopify sends stores on third-party themes to the theme developer, a Shopify web performance expert or a Shopify Partner (Shopify Help Center, October 2026).
- Interaction to Next Paint (INP) is the failing metric. Interaction delay comes from JavaScript across theme and apps at once, which is hard to attribute without profiling.
- The estate is bigger than one theme. Multiple themes, a headless front end or several agencies adding tags need governance, not a single fix.
- Regressions keep coming back. Deploi's script governance record explains why: app-injected scripts "land in the theme without a deploy," so a pipeline check alone misses them (October 2026).
How an agency structures a paid app-stack audit lists what its deliverable must contain.
When should you hire nobody?
You should hire nobody when your field metrics sit inside the Good bands, because no shopper-facing problem is left to fix.
Deploi’s point of view
Our own position, from building on Shopify. Kept apart from the sourced facts above.
- Where we disagree
- The standard agency pitch is a full performance audit as the entry point. For a store that has never done the app embed inventory, that is selling week three before week one. We would rather hand a client the four-item checklist first, even when it costs us the first piece of work, and be right about what is left.
- What we fix first, in our judgment
- In our judgment, the fix that moves field Largest Contentful Paint is usually narrow: the hero image, the font loading strategy and two or three scripts moved out of the critical path. The wide rewrites proposed at kickoff rarely earn out, and in widget work the fix is often a conversation with the vendor rather than a commit. That is our view from client work, not a measured result.
- What this page adds
- the symptom-to-owner triage, the four in-house fixes to run first, and the field-data thresholds that say the problem is already solved.
Decision records behind this answer
When this answer changes
- Shopify changes its web performance summary's 30-day window, the reports' 90-day retention, the 36-hour delay or the Good thresholds.
- Google changes a Core Web Vital or its Good threshold.
- Shopify adds per-app or per-script attribution to its web performance reports.
- Shopify Support starts helping with performance on third-party themes.
- Hyperspeed changes its $59 price, or Shopify ships native script deferral.
Related questions
This follows from Is upgrading to Shopify Plus actually a legitimate performance-improvement strategy, or does it only address checkout-specific scale, not general site speed?
- How do agencies typically structure a paid app-stack audit engagement, and what should be in the deliverable?
A paid app-stack audit, in Deploi's structure, delivers four artifacts: an app inventory with spend, a script-to-app attribution map, a ranked remove, replace or keep list, and a written install policy.
- Do all Shopify stores on the same plan get identical baseline infrastructure performance, or does Plus actually run on faster infrastructure?
Shopify Plus runs on the same storefront infrastructure as every plan, because every online store gets fast global servers and a Cloudflare content delivery network (Shopify Help Center, October 2026).
Sources
Every figure on this page traces to one of these. Official and vendor pages are checked on the date shown; Deploi records are our own published decisions.
- Shopify Help Center, Improving your online store performance
- Shopify Help Center, Web performance reports
- web.dev, Web Vitals
- web.dev, Optimize Largest Contentful Paint
- Shopify Help Center, Overview of web performance
- Rvere, Hyperspeed EXTREME Page Speed on the Shopify App Store
- Deploi Build vs. Buy, Build or Buy Performance Monitoring & App Audits on Shopify?
- Deploi Build vs. Buy, Build or Buy Site Speed Optimization on Shopify?
- Deploi Build vs. Buy, Build or Buy Third-Party Script Governance on Shopify?