← Back to memo

Methodology

How this data was collected, filtered, and categorized. No internal Walmart data is used anywhere on this site — everything below is sourced from public pages.

Scope

Brands covered: Walmart Auto Care Centers, Sam's Club Tire & Battery, Costco Tire Center, Discount Tire. Costco and Discount Tire are included as benchmarks to ground “what good sounds like” in customer voice — not as a competitive analysis.

Journey stages (exactly one assigned per quote):

  1. Discover — browsing tires, fitment, pricing on walmart.com or the Walmart app
  2. Schedule — booking the appointment, picking service, time, and store
  3. Check-in — arrival, key handoff, wait expectations
  4. Service — the work itself, communication during, verification
  5. Pickup — payment, confirming work done, walking out
  6. Post-visit — warranty, rotation reminders, rebooking, returns

Data sources

40 verbatim customer quotes from public sources, collected April 10–14, 2026. 18-month freshness window (October 2024 – April 2026) where dates are visible.

SourceQuotes
Reddit40
Total40

Split: 24 Walmart/Sam's quotes, 16Costco/Discount Tire benchmark quotes. v1 is Reddit-only because that's what I could collect verbatim and date-stamped in the time I gave myself. v2 would fold in ConsumerAffairs, Google/Yelp store reviews, Walmart + Sam's app store reviews, and YouTube comments — each from a different failure mode distribution, which is the point of adding them.

Inclusion rule

A friction theme only surfaces in the top-level memo if it appears in ≥ 2 independent Reddit comments— different threads, different users — for the same brand and the same journey stage. One viral thread doesn't make a theme. v2's corroboration rule tightens to “≥ 2 independent platforms” once non-Reddit sources are added.

Quote selection was human review, not LLM-judged. I read the post, made sure it was (a) about the right brand, (b) about the right journey stage, (c) specific enough to teach something (“wait was long” got cut; “booked at 9am, walked out at 1pm” stayed).

What this can and can't say

Can say: which journey stages generate the most specific, repeatable public complaints, with verbatim examples, and what the comparable positive voice sounds like for competitors.

Cannot say: population-level satisfaction, incidence rate, or how these patterns would rank against internal CSAT / NPS data the Walmart team already has. Public review data skews toward customers with strong feelings — usually negative — so absolute shares should be read as relative-within-corpus, not absolute-in-the-world.

What v2 would add

v1 is intentionally small and Reddit-only — the fastest way to prove the journey framing works before investing in a broader pipeline. v2 sharpens three things.

1 · More sources, different failure modes

Reddit captures a specific type of customer: the one who wanted to tell a story. Each additional source adds a different lens — and the blind spots of the current data are where the most expensive leaks tend to live.

SourceWhat it catches that Reddit doesn't
App store reviewsPre-visit abandoners — the shoppers who churn at the Discover / Schedule stages, never enter a store, and never post about it.
Google / Yelp store reviewsStore-level variance. Friction is geographic — some ACCs are great, some aren't. Per-store review data would show which patterns are systemic vs. hot-spot.
ConsumerAffairsEscalation-stage complaints — warranty denials, damage claims, formal disputes. The failure modes that cost money on the back end.
YouTube commentsShoppers actively comparing ACC to Costco / Discount mid-decision — the Discover-stage switchers.

Corroboration rule tightens with the source expansion: v2 requires ≥ 2 independent platforms (not just 2 Reddit comments) before a theme lands on the memo.

2 · Weekly auto-refresh

Replace the manual curation with a scheduled job that pulls, dedupes, and re-tags on a weekly cadence. Pattern already running on two of my other research projects (tinker-flywheel, managed-agents-pulse) — a day of work to port, and it keeps the link live for anyone reopening it weeks from now.

3 · Per-store heatmap

Once Google review data is in, every quote gets a store ID. A map view would show which ACC locations generate disproportionate friction at each journey stage — turning “check-in is broken” into “check-in is broken at these 40 stores, where the common thread is [X].” That's the version a regional ops lead could actually act on.