Build vs. Buy>Shipping, Delivery & Fulfillment>Shipping protection / insurance

Should You Build or Buy Shipping Protection on Shopify?

Written by Deploi EditorialReviewed by Martin Dejnicki, Director of SEO & AI SearchUpdated August 2026Pricing verification pending

Shipping protection is a self-insure build for most mid-market Shopify stores past roughly 5,000 orders a month: a cart-transform Function charges your own fee, claims get paid from the pool, and the spread between fees collected and claims paid out stays on your P&L instead of a vendor's (illustrative math; typical loss rates run 1–2%). Buy insurer-backed only for zero claims ops. The quiet third option: charge nothing and cover reorders as policy.

Your profile — see how the verdict shifts

VerdictBUILD self-insurance (at scale) · BUY for zero claims ops · or charge nothing
Buy score
5.4
Build score
7.6
Confidence
MediumSelf-insure math wins at typical loss rates, but your own claims data decides it; at low loss rates the honest verdict is to skip the fee entirely and cover reorders
Reference scenario
$20M–$100M GMV · 8,000 orders/mo · agency dev bench · single storefront
As of
August 2026

Decision at a Glance

Your profileVerdictWhy
Under 1,000 orders/moWAITClaims are rare enough to handle by hand. Reorder the lost package, keep the goodwill, and skip the checkout optics of a fee; a protection line isn't where this budget goes.
1,000 – 5,000 orders/moDEPENDSRun a quarter of loss data first. Insurer-backed gets you the toggle with zero ops; if claims sit under about 1% of orders, covering reorders as policy still beats both paths.
5,000 – 20,000 orders/moBUILDThe fee pool now clears claims with real margin at typical loss rates (illustrative math), and the spread a revenue-share app would keep becomes your line. Claims ops is a part-time job, not a department.
20,000+ orders/moBUILDAt this volume the spread is a budget line and the loss data is carrier-negotiation leverage. The claims desk professionalizes either way; better on your side of the ledger.

What Shipping protection / insurance Actually Drives

OutcomeImpactHow it works
Revenue — directHighThe familiar 'protect my order for $2.98' toggle (illustrative fee) collects more than claims pay out at typical loss rates; self-insured, that spread lands on your P&L instead of a vendor's.
Customer experienceHighThe claim moment decides everything: an instant reshipment turns a stolen package into a loyalty story, while a third party's denial email lands as your one-star review.
Retention & LTVMediumHow the worst delivery day gets handled is the strongest second-order signal a store sends; reshipment-first policies convert the angriest moment into the most-told story.
Data & insightMediumClaims tagged by carrier, lane, and SKU show where packages actually die, feeding packaging fixes and carrier negotiations most stores run on anecdote instead.
Operational efficiencyMediumA rules-based claims flow with an evidence bar, auto-approve threshold, and reshipment trigger replaces the ad-hoc support ping-pong that lost-package emails create today.

Spend ceiling: Size everything against your true loss line, not the fee's revenue potential. If lost-and-damaged runs under roughly half a percent of orders, the honest ceiling is zero: skip the toggle, cover reorders as policy, and bank the checkout trust instead.

What buying enables (top apps)

  • + Live this week: toggle, claims portal, and a staffed claims desk you never build
  • + Licensed-insurer paper answers the 'is this insurance?' question so you don't have to
  • + Fraud screening and claim SLAs handled at a scale no single store matches
  • + Package tracking and delivery notifications often ride along in the same install

What building additionally unlocks

  • + The spread itself: fees minus claims becomes a margin line instead of a vendor's business model
  • + Pricing freedom a revenue-share model can't offer: paid toggle, free perk above a cart threshold, loyalty benefit, or no fee at all
  • + Claim rules and reshipment-first UX inside your own helpdesk, tuned to your fraud reality
  • + Loss data by carrier, lane, and SKU owned outright for packaging and carrier negotiations

Find Your Verdict in 3 Questions

  1. Do lost, damaged, or stolen packages run above roughly 1% of your orders?

    Yes: Go to question 2.

    No: Your verdict: WAIT — skip the toggle, cover the rare reorder as CX policy, and re-run the math if the loss line grows.

  2. Do you want the protection fee to become your margin rather than a vendor's revenue share?

    Yes: Go to question 3.

    No: Your verdict: BUY — an insurer-backed app delivers the toggle and a staffed claims desk this week; the spread is the price.

  3. Can support own a claims workflow, and can your dev bench take a 4–8 week build?

    Yes: Your verdict: BUILD — self-insure with a cart-transform Function and keep the spread, the rules, and the loss data.

    No: Your verdict: BUY for now — collect a year of attach and loss data on the app, then re-decide with your own claims math.

The TCC Scorecard — 12 Dimensions

TCC — Total Cost of Capability: what it actually costs to have this capability over three years, whichever way you get it. Each dimension is scored 0–5 for both paths. How we score →

DimensionBuyBuildWhy
Cost
Acquisition & implementationAn insurer-backed app is live in days with the claims desk included; the build is an estimated 4–8 weeks of Function, toggle, and claims flow (Deploi estimate, illustrative).
Recurring feesNo subscription on the app path, but the fee pool and its spread exit to the vendor forever; the build keeps the pool and pays claims plus modest upkeep out of it.
Maintenance & upgradesThe vendor maintains the toggle through checkout changes; a build owns Function and Checkout UI extension version bumps (~$2,500–$7,000/yr, Deploi estimate, illustrative).
Switching & exitUninstalling the app is easy, but the claims history that would price your self-insure pool leaves with it; turning off your own Function strands nothing.
Risk
Vendor riskYour protection promise rides the vendor's claim decisions and model changes, and protection economics invite repricing; a build has no vendor to lose.
Security & compliance surfaceInsurer-backed apps carry licensed paper; a self-insure build must be worded as a reshipment guarantee, not insurance, and that wording deserves counsel review before launch.
Platform-deprecation exposureBoth paths live on the sanctioned checkout stack now that Scripts stopped executing June 30, 2026 (July 2026 research); a fresh Functions build starts clean with no legacy surface to migrate.
Value
Fit to requirementApps ship their toggle, their fee logic, their claim rules; a build sets your fee, your evidence bar, your reshipment-first policy, and can even price at zero.
Time to marketThis week with a staffed claims desk versus one to two months building your own.
Performance & scaleA cart-transform Function runs server-side with no injected checkout widget, and the economics improve with volume instead of scaling a vendor's share.
Data ownership & AI-readinessClaims tagged by carrier, lane, and SKU are packaging and carrier-negotiation intelligence; owned, they feed your reporting and future pricing instead of sitting in a vendor portal.
Focus & opportunity costA claims desk is real, undifferentiated work someone must own; the build earns its roadmap slot only when volume makes the spread loud.

The App Landscape

AppStatusPricingBest for
RouteLiveThe insurer-backed category's best-known name: customer-paid toggle, claims portal, and a claims desk you never staffCustomer-funded per-order fees in a $1–$5 band, revenue-share economics (illustrative)Zero-ops protection live this week, with licensed-insurer paper behind the promise
RedoLiveNewer entrant pairing returns with checkout-funded coverageShopper-funded coverage model rather than flat SaaSFolding protection into a broader returns offer under one vendor

The Build Path

  • Cart-transform Function + checkout toggle: A Shopify Function adds the protection fee as a line item and a Checkout UI extension renders the opt-in toggle: the sanctioned stack since Scripts stopped executing June 30, 2026 (July 2026 research).
  • Claims workflow in your helpdesk: A claim form feeds your existing support tooling: evidence rules, an auto-approve threshold for small claims, reshipment via draft order or store credit, and Shopify Flow handling the routing.
  • The pool ledger: A monthly report of fees collected versus claims paid: the number that justifies the program to finance and, tagged by carrier and lane, the dataset that drives packaging fixes.
Effort band
$15,000–$35,000 build (Deploi estimate, illustrative); lean scopes land in the $10–25K contact-form band, claims-automation depth pushes toward $25–75K
Typical timeline
4–8 weeks (Deploi estimate, illustrative)
Maintenance, honestly
~15–20% of build cost per year, roughly $2,500–$7,000/yr (Deploi estimate, illustrative): Function and Checkout UI extension API version bumps (~every six months), claim-rule tuning, and toggle UX tweaks. Claims payouts come from the fee pool, not this line.
What you own — and what you take on
You own: the fee, the pool, the claim rules, the loss data, and full pricing freedom, including charging nothing. You take on: the claims desk, the fraud judgment calls, and the guarantee wording that keeps you out of regulated-insurance territory.

3-Year Total Cost of Capability

Buy (app path)Build (custom path)
Year 0 (setup)$0 (customer-funded)$15,000–$35,000
Years 1–3 (recurring)$0 cash; the pool and its spread exit with the vendor$7,500–$21,000 upkeep; claims paid from the pool
3-year total$0 cash · ≈$60,000–$70,000 spread foregone (illustrative)≈$22,500–$56,000, with the same spread retained
Illustrative cumulative cost over 36 months$0$11k$21k$32k$43kMo 0Mo 12Mo 24Mo 36Buy (app path)Build (custom path)
Illustrative and deliberately upside-down: the app line sits at zero because customers fund it; the real cost is the roughly $1,900-a-month spread (illustrative) that leaves with the vendor in the reference scenario. Credit that spread to the build and it pays back around month 14. Claims volatility is why you keep a pool buffer; your loss rate decides everything.
  • All figures illustrative samples for the reference scenario — not quotes, not verified pricing.
  • App path shown at zero cash cost: customers fund it, the vendor keeps the pool, and any revenue share back to the store is ignored.
  • Build path: 8,000 orders/mo, 55% attach at a $1.85 fee, 1.2% loss rate at a $65 average reorder (the FAQ math); three-year horizon.

What the Sticker Price Hides

On the buy path

  • Revenue-share economics never sunset: the spread between customer-paid fees and claims paid out is the vendor's margin, and it grows with your order volume (community-reported pattern)
  • The vendor's claim denial reads as your brand's refusal — customers review the store, not the insurer
  • Checkout optics are yours to defend: pre-selected or aggressively nudged toggles have drawn merchant and customer pushback (community-reported pattern)
  • Your claims and loss-rate history accumulates in the vendor's portal: the exact dataset you'd need to self-insure later

On the build path

  • The claims desk is real work: evidence rules, fraud judgment calls, and reshipment SLAs land on your support team
  • Wording discipline is non-negotiable: sell a reshipment guarantee, not 'insurance', and get counsel review before launch
  • A carrier meltdown or porch-piracy spike hits your pool, not an insurer's balance sheet; keep a buffer
  • ~$2,500–$7,000/yr upkeep, about 15–20% of build cost per year (Deploi estimate, illustrative)

What Merchants Say

Protection apps draw claim-denial blowback: the toggle collects at checkout, but a denied or slow claim lands as a one-star review on the store, not on the insurer behind it.
app-store 1–2★ review theme
The recurring self-insure thread: merchants who ran a year of claims numbers report the fee pool clearing payouts with room to spare, and regret leaving the spread with a vendor so long.
community-reported pattern

If You Change Your Mind Later

If you bought and outgrow it

Mechanically painless: uninstall and the toggle disappears. What leaves is the claims and loss-rate history that would have priced your self-insure pool, so export whatever your tier allows before you go. Expect reorder requests to re-route to your own support desk, and decide in advance whether the protection promise quietly continues as policy.

If you built and want out

Reversible by design: switch off the Function and the fee stops that day. The claims workflow keeps working as a plain CX guarantee, the pool ledger and loss data stay yours, and retreating to an insurer-backed app later means arriving with a year of your own claims math instead of a vendor's word.

When This Answer Changes

We're watching for:

  • Shopify shipping a native delivery-guarantee or protection primitive at checkout (none as of July 2026 research)
  • Regulatory or state-level attention to how retail 'shipping protection' fees are framed
  • Route-class model or packaging changes in the protection category (re-verify quarterly)

Verdict change log:

No changes since first publication (August 2026).

Common Questions

Is it legal to charge our own shipping protection fee?

Generally, merchants run self-insured protection as a reshipment guarantee, a service you deliver, rather than insurance, which is a regulated product only licensed carriers can sell. The wording is the whole game: 'we'll replace it' is a service promise, while 'coverage' and 'policy' drift toward regulated territory. Get counsel to review your language before launch; insurer-backed apps exist partly to carry that paper for you.

What does self-insured shipping protection actually net?

Run your own numbers, but here's the illustrative shape: at 8,000 orders a month with a 55% attach rate on a $1.85 fee, the pool collects about $98,000 a year; claims at a 1.2% loss rate and a $65 average reorder cost pay out roughly $75,000 (all figures illustrative). The difference is the spread — the exact margin the revenue-share model keeps. Your loss rate moves this more than any other input.

What if we'd rather not charge customers anything?

Self-insuring reorders is often the right answer, and nobody selling protection will suggest it. If your loss rate is low, quietly covering reorders costs less than the checkout optics of a fee, and 'if it doesn't arrive, we replace it, free' is a conversion asset, not a cost center. You still want the build's claims workflow (rules, evidence, reshipment triggers); you just skip the fee. Run the math both ways before deciding.

Your Next Steps

If you're going with BUILD(matches your selected profile)

  1. Pull 12 months of lost, damaged, and stolen reships and compute your true loss rate: the whole decision is this number
  2. Model the pool: attach rate times fee versus claims times average reorder cost, labeled illustrative until measured
  3. Scope the cart-transform Function, the Checkout UI extension toggle, and the claims flow in your helpdesk
  4. Set claim rules up front: evidence bar, auto-approve threshold, reshipment-first
  5. Get counsel eyes on the guarantee wording before launch: a service promise, not insurance

If you're going with BUY

  1. Shortlist insurer-backed apps and read the claim-denial reviews before the feature list
  2. Decide in advance who fronts a denied claim: your policy for when the vendor says no
  3. Check what claims and loss data you can export on your tier; you'll want it to self-insure later
  4. Diary a re-decision at 12 months with your own attach and loss numbers in hand

Official Docs & Sources

Official documentation linked for verification — our verdicts and estimates are our own.

Ready to keep the spread?

We build the full self-insure loop: checkout Function, protection toggle, claims workflow in your helpdesk, and the pool ledger finance will ask for, sized against your real loss rate. And if your math says charge nothing, we'll say so.

Contact us today

Ecommerce development at Deploi

Verdict scored for the reference scenario above. Estimates are not quotes; app pricing is banded from public listings. The claims math is illustrative, and your measured loss rate decides this category. Full scoring anchors: see the TCC methodology.

Read how we score these decisions (the TCC Framework). No affiliate links, no paid placement — no app vendor pays to appear here.

No affiliate links. No paid placement. We make money building and integrating solutions — not on referral fees.