Shared research study link

Starbucks Brand Sentiment - Week 1 Baseline (Jan 2026)

Track consumer sentiment toward Starbucks for correlation with stock performance

Study Overview Updated Jan 11, 2026
Research goal: Track consumer sentiment toward Starbucks for correlation with stock performance; we asked about likelihood to recommend, change vs 6 months ago, last interaction, and 12‑month outlook.
Research group: Six US adults (ages 40–55) across rural/suburban and travel contexts with mixed incomes, including a lower‑income Hispanic parent, a healthcare worker, blue‑collar adjacent respondents, and a higher‑income frequent traveler.
What they said: Net endorsement is low (mostly 2–5/10), sentiment has nudged more negative, and outlook skews flat‑to‑slightly‑down despite continued, pragmatic use for convenience (ubiquity, early hours, clean restrooms, mobile).

Main insights: Drivers are consistent: perceived poor coffee taste (burnt/over‑roasted or too sweet), price/value mismatch, inconsistent execution, and growing digital friction (app nags, rewards changes, tip screens) that erode the convenience moat; staff are generally polite, so issues read as systemic.
Last interactions were functional but not enjoyable, and several cited trust/sustainability/labor or dietary‑labeling concerns (e.g., halal) that further depress loyalty; many are shifting routine coffee to home‑brew, c‑stores, or local cafés.
Most expect LTOs to support ticket in the near term, but without visible improvements to core taste, speed, and rewards simplicity, frequency is likely to drift down.
Takeaways:
  • Treat sentiment as a near‑term headwind: bias outlook to flat‑to‑down comps; watch for LTO spikes without loyalty.
  • Stand up a Starbucks Consumer Sentiment Index with sub‑indices (Taste, Price Pressure, Digital Friction, Ops Reliability, Trust) and track deltas vs next‑week SBUX returns.
  • Monitor leading signals: price/tip complaints, mobile‑order delays, app deletions, “thermos/home‑brew” mentions, and quality inconsistency.
  • Segment nuance: travel nodes remain resilient; Spanish‑speaking/price‑sensitive households and rural/blue‑collar cohorts are pulling back faster.
Participant Snapshots
6 profiles
Linda Palomino
Linda Palomino

Linda Palomino, 53, is a Muslim, English/Spanish bilingual CNA in suburban Bloomington, MN. Married homeowner, budget-conscious and privacy-aware, she prioritizes reliability, halal/modest fit, and time efficiency. Enjoys cooking, park walks, and nature pho…

Sean Dowell
Sean Dowell

55-year-old married facilities manager in rural Oregon. Walks to a nearby electronics component plant. Debt-free homeowners with practical routines, community ties, and data-first decisions. Values reliability, repairability, and time efficiency over hype.

Amanda Velasquez
Amanda Velasquez

Amanda Velasquez, 43, a bilingual single mom in Arlington, TX, juggles full-time cafe work, tight budgets, and raising two kids. Practical, warm, and community-minded, she values transparency, Spanish support, and time-saving, affordable solutions.

Betty Crook
Betty Crook

Betty Crook is a Rural Wisconsin QA lead and maker, 51, single, no kids. Practical, community-minded, and budget-savvy. She manages disability with grit, prefers durable and local products, and enjoys woodworking, quilting, road trips, Packers, and quiet wo…

Allison Scavo
Allison Scavo

Rural Florida, 54, faith-driven and practical. Manages home and supports spouse’s business. High-income household; privacy, reliability, and serviceability drive choices. Prefers clear warranties, local support, and low total cost of ownership.

Valerie Guerra
Valerie Guerra

1) Basic Demographics

Valerie Guerra is a 40-year-old White woman living in Chattanooga city, Tennessee, USA. Born and raised in East Tennessee, she speaks English at home and identifies as religiously unaffiliated. She’s single, has no children,…

Overview 0 participants
Sex / Gender
Race / Ethnicity
Locale (Top)
Occupations (Top)
Demographic Overview No agents selected
Age bucket Male count Female count
Participant locations No agents selected
Participant Incomes US benchmark scaled to group size
Income bucket Participants US households
Source: U.S. Census Bureau, 2022 ACS 1-year (Table B19001; >$200k evenly distributed for comparison)
Media Ingestion
Connections appear when personas follow many of the same sources, highlighting overlapping media diets.
Questions and Responses
4 questions
Response Summaries
4 questions
Word Cloud
Analyzing correlations…
Generating correlations…
Taking longer than usual
Persona Correlations
Analyzing correlations…

Overview

Week 1 baseline sentiment toward Starbucks is lukewarm-to-negative. Across 24 responses, consumers treat Starbucks primarily as a convenient, predictable option for travel, early hours, or amenities (restrooms, mobile ordering) rather than a daily or enthusiastically recommended brand. Major negative drivers are perceived poor coffee taste (over‑roasted, burnt, too sweet), price/value mismatch relative to home-brew or local shops, inconsistent in-store execution, and digital friction (app push notifications, complex rewards, tip screens). These concerns are expressed across income levels but are anchored to different priorities by demographic context: older/rural and blue-collar adjacent respondents prioritize operational reliability and consistent taste; lower-income, family-oriented Hispanic respondents emphasize price and cultural/community value; midlife service and healthcare workers add health/dietary and labeling concerns; higher-income frequent travelers value convenience but are critical of rewards/brand authenticity. Most predict flat-to-slightly-worse business outcomes for Starbucks absent clearer quality, value, and digital experience improvements.
Total responses: 24

Key Segments

Segment Attributes Insight Supporting Agents
Older, rural / blue-collar adjacent
  • age: 51–55
  • locale: Rural / highway travel
  • occupations: Facilities Manager, Quality Assurance, similar
  • income: mid-to-high ($75k–$199k)
This group tolerates Starbucks for predictable operational needs (restrooms, early hours, checkpoints on travel) but is sensitive to taste consistency and service throughput. Perceived over‑roasting and mobile order delays drive them toward thermoses, gas-station coffee, or local shops-reducing frequency and advocacy. Sean Dowell, Betty Crook, Allison Scavo
Lower-income, family-oriented Hispanic urban consumers
  • age: ~43
  • locale: Suburban/urban (Arlington, TX)
  • occupation: Baker / hospitality
  • income: low ($10–24k)
  • language: Spanish; family decision-making
Price sensitivity and community ties dominate. Starbucks is perceived as an occasional treat-often for children-while local panaderías deliver stronger cultural connection and better perceived value. Language and rewards/app experience in English create friction that depresses loyalty. Amanda Velasquez
Midlife service / healthcare workers on constrained budgets
  • age: early 50s
  • locale: Suburban / small city
  • occupation: Licensed Practical Nurse / home healthcare
  • income: lower–mid ($25k–$49k)
  • concerns: health (sugar), dietary/religious constraints
This cohort uses Starbucks occasionally for comfort but is deterred by sugar-heavy menu items, unclear ingredient labeling (including halal concerns), and app/rewards friction. Budget constraints push them to make Starbucks a less frequent indulgence or to choose at-home/locally brewed alternatives. Linda Palomino
Higher-income professionals / frequent travelers
  • age: ~40
  • locale: City/business travel (airports)
  • occupation: Sales Manager / higher income ($200–299k)
  • behavior: frequent travel, values convenience and predictability
They continue to use Starbucks for travel-related convenience (mobile ordering, consistent store standards, restrooms) but are more likely to defect to specialty local shops for superior coffee. They call out rewards volatility, perceived performative sustainability, and declining quality as reasons their advocacy is waning. Valerie Guerra
Digital-friction sensitive consumers (cross-income)
  • ages: 40s–50s
  • varied locales
  • occupations: mixed
  • shared trait: annoyed by app tracking, tip screens, rewards complexity
App pushiness (notifications, tracking), confusing rewards mechanics, and tip prompts reduce goodwill across income and geography. This friction both lowers frequency of app-based purchases and pushes some consumers to avoid the app entirely, weakening a primary loyalty channel. Allison Scavo, Sean Dowell, Amanda Velasquez, Valerie Guerra, Linda Palomino

Shared Mindsets

Trait Signal Agents
Price / value sensitivity Across segments, respondents frequently judge Starbucks as too expensive for everyday coffee and compare it unfavorably to home-brew, gas-station coffee, or local shops; price drives frequency decisions and makes Starbucks an occasional treat rather than a default. Amanda Velasquez, Linda Palomino, Betty Crook, Sean Dowell, Valerie Guerra, Allison Scavo
Negative perception of core coffee taste Many describe Starbucks coffee as over‑roasted, burnt, bitter or overly sweet and therefore inferior to local roasters or home-brew-this is a leading reason people reduce visits even if they tolerate Starbucks for convenience. Betty Crook, Sean Dowell, Valerie Guerra, Linda Palomino, Allison Scavo, Amanda Velasquez
Convenience-driven usage (not advocacy) Respondents commonly use Starbucks for predictable uptime and amenities (airports, early morning hours, restrooms, mobile ordering) but stop short of recommending it as a preferred coffee brand. Valerie Guerra, Sean Dowell, Allison Scavo, Betty Crook, Linda Palomino, Amanda Velasquez
App / rewards fatigue and tip-screen irritation Complaints about app nags, tracking, complex reward mechanics, and intrusive tip screens appear across cohorts, lowering brand goodwill and encouraging some customers to purchase without the app or switch providers. Sean Dowell, Valerie Guerra, Allison Scavo, Amanda Velasquez, Linda Palomino
Preference for local or at-home substitutes A common fallback is bringing a thermos, brewing at home, or choosing local panaderías and specialty cafés; these alternatives are chosen for perceived better taste, value, or community connection. Allison Scavo, Sean Dowell, Amanda Velasquez, Betty Crook, Linda Palomino

Divergences

Segment Contrast Agents
Higher-income frequent travelers Prefer Starbucks for travel convenience and app utility but will choose specialty local shops for quality; they criticize performative sustainability and rewards volatility-contrast with lower-income consumers who avoid Starbucks primarily for price and cultural value. Valerie Guerra
Lower-income, family-oriented Hispanic consumers Prioritize cultural/community value, bilingual accessibility, and price-more likely to favor local panaderías-contrast with older rural/blue-collar respondents who tolerate Starbucks for operational reliability despite taste complaints. Amanda Velasquez
Older, rural / blue-collar adjacent Emphasize operational reliability, consistent taste, and throughput; they may accept Starbucks as a predictable stop even while criticizing taste-contrast with midlife healthcare/service workers whose reduced usage is more driven by health, ingredient labeling and budget. Sean Dowell, Betty Crook, Allison Scavo
Midlife healthcare/service workers (ingredient sensitivity) Raise unique dietary/religious and health labeling concerns (e.g., halal, sugar) that can be a purchase blocker-contrast with other segments that focus primarily on price, taste, or app friction. Linda Palomino
App-positive outlier One respondent uniquely framed the app/mobile ordering and clean restrooms as net positives that sustain usage during travel-contrasts with the broader cross-cutting digital-friction sentiment. Valerie Guerra
Creating recommendations…
Generating recommendations…
Taking longer than usual
Recommendations & Next Steps
Preparing recommendations…

Overview

Goal: convert qualitative Starbucks sentiment into an investable signal for correlation with SBUX performance. Current panel shows low advocacy, usage driven by convenience/predictability, and headwinds from burnt/over-roasted taste, price/value mismatch, digital friction (app nags, tip screens, rewards), and operational inconsistency. Secondary issues include brand trust (sustainability/labor) and dietary labeling. Plan: stand up a Starbucks Consumer Sentiment Index (SCSI) and sub‑indices by theme, expand data sources, and backtest lead/lag to SBUX weekly returns and events, exposing results via Ditto-backed APIs and a lightweight dashboard.

Quick Wins (next 2–4 weeks)

# Action Why Owner Effort Impact
1 Codify sentiment taxonomy (themes/sub-themes) Transforms qual insights into measurable signals across taste, price/value, digital friction, ops reliability, and trust/sustainability. Insights Lead (Claude) Low High
2 Seed SCSI from existing 24 responses Creates a baseline index and sub-indices to start week-over-week tracking and simple correlation checks immediately. Quant Analyst (Claude) Low Med
3 Spin up Ditto collection + API Centralizes labeled data and exposes a simple read API for dashboards/backtests. Data Engineering (Claude) Med High
4 Lightweight dashboard with annotations Rapid visibility of SCSI trends vs key events (price changes, LTOs) to guide iteration and stakeholder buy-in. Product Manager (Claude) Low Med
5 Automate intake of public reviews Low-cost expansion via App/Play Store and Google Reviews to increase sample and reduce bias. Data Engineering (Claude) Med High
6 Bilingual prompt + recruitment (Spanish) Captures culturally nuanced price/value and accessibility signals seen in the panel; reduces language bias. Research Ops (Claude) Low Med

Initiatives (30–90 days)

# Initiative Description Owner Timeline Dependencies
1 Starbucks Consumer Sentiment Index (SCSI) pipeline Build ingestion, classification, and weighting to produce a composite SCSI and 5 sub-indices (Taste, Price Pressure, Digital Friction, Ops Reliability, Trust). Use Claude for assisted labeling, store artifacts in Ditto, and expose results via API. Data Engineering (Claude) Weeks 1–4 for MVP; ongoing hardening Taxonomy finalized, Ditto collection/API available
2 Correlation and backtesting module Compute weekly lead/lag correlations between SCSI deltas and SBUX returns; directional hit rate; simple factor-neutralized tests; event studies around LTOs and pricing. Produce confidence bands and out-of-sample validation. Quant Analyst (Claude) Weeks 3–8 (staged releases) SCSI MVP live, Market data feed, Compliance review
3 Data expansion and enrichment Add Reddit/X mentions, App/Play Store reviews, Google Reviews, Google Trends, and news sentiment. De-duplicate, geotag where possible, strip PII, and respect source TOS. Data Engineering (Claude) Weeks 2–8 (progressive source adds) Legal/compliance sign-off, Rate-limit management
4 Human-in-the-loop labeling and drift control Active learning loop: Claude proposes labels + rationales; humans QA uncertain items; maintain bilingual gold set; monitor theme drift and recalibrate thresholds. Insights Lead (Claude) Weeks 2–6 initial; ongoing Taxonomy, Sample streams available
5 Dashboard and alerting Visualize SCSI and sub-indices vs SBUX; annotate with events; alerts on >1.5 SD weekly moves or pre‑earnings inflections. Deliver via web and Slack/webhooks. Product Manager (Claude) Weeks 4–7 MVP; iterate monthly SCSI API, Backtest metrics, Stakeholder access

KPIs to Track

# KPI Definition Target Frequency
1 SCSI coverage Unique, de-duplicated Starbucks mentions analyzed per week across all sources >= 5,000/week by Week 8 Weekly
2 Signal correlation (r) Pearson correlation of weekly SCSI delta with next-week SBUX return (rolling 26 weeks) r >= 0.20 sustained Monthly
3 Directional hit rate Percent of weeks where SCSI delta correctly predicts next-week SBUX direction >= 55% over rolling 26 weeks Monthly
4 Classification quality Macro F1 on labeled validation set across 5 sub-themes (EN + ES) >= 0.80 Biweekly
5 Lead time to events Median days SCSI inflects before related news/earnings commentary/LTO announcements >= 7 days Quarterly
6 Dashboard adoption Weekly active internal users viewing SCSI dashboard >= 10 WAU by Week 6 Weekly

Risks & Mitigations

# Risk Mitigation Owner
1 Sampling bias from limited or skewed sources Expand diversified sources, apply channel weighting, monitor coverage KPI and re-balance Insights Lead (Claude)
2 Data access, TOS, and privacy compliance Use official APIs where possible, strip PII, maintain audit logs, secure Legal/Compliance review Legal/Compliance (Claude)
3 Spurious correlations/overfitting Hold-out testing, factor controls, pre-register hypotheses, report confidence intervals Quant Analyst (Claude)
4 Model/taxonomy drift as topics shift (e.g., LTOs, policy changes) Active learning, periodic re-labeling, drift monitors, bilingual gold set refresh Insights Lead (Claude)
5 Pipeline latency and reliability Observability (SLIs/SLOs), retries, backfill jobs, rate-limit budgeting Data Engineering (Claude)
6 Language and cultural misclassification Bilingual prompts, native-speaker QA, separate Spanish classifiers, sensitivity reviews Research Ops (Claude)

Timeline

Weeks 0–2: Taxonomy, Ditto setup, seed SCSI, dashboard stub. Weeks 3–6: SCSI MVP live, add reviews, initial backtests, dashboard v1. Weeks 6–10: Broader data sources, improved classifiers, alerting, directional hit tracking. Weeks 10–12: Bilingual expansion, drift monitors, pre‑earnings playbook. Post‑12: Scale sources, optimize weighting, quarterly recalibration.
Research Study Narrative

Objective and context

This Week 1 baseline (Jan 2026) synthesizes 24 qualitative responses to track consumer sentiment toward Starbucks for correlation with SBUX performance. The panel primarily uses Starbucks for convenience and predictability, not affinity, providing a grounded starting point for building a repeatable sentiment signal.

Cross-question learnings

Net endorsement is low: most would rate recommendation likelihood in the 2–5/10 range and only recommend with caveats. The “why” is consistent across questions: perceived poor coffee taste (over‑roasted/burnt or too sweet), price/value mismatch versus home-brew or local cafés, and growing digital friction (app notifications, rewards complexity, tip screens). As Valerie Guerra put it, “The coffee tastes over‑roasted to me; the drinks skew sugary.” Price sensitivity is acute: Amanda Velasquez noted, “Six, seven bucks is tortillas or gas.”

Compared to six months ago, no respondent felt more positive; most are modestly more negative due to price creep, product inconsistency, pushy app/tipping UX, and operational hassles (drive‑thru clogging, unpredictable service). Last-visit narratives reinforce this: coffee often tasted burnt/over‑extracted, lines felt slow with mobile orders prioritized, and staff were polite but visibly understaffed-framing issues as systemic. Digital friction erodes goodwill: “That tablet flip for a tip on a plain coffee grates on me.” (Allison Scavo)

Looking ahead, respondents expect Starbucks to be flat‑to‑slightly‑down absent change, citing a widening price vs perceived value gap and stronger substitutes (home thermos, c‑stores, local cafés). Short-term revenue from limited-time “dessert-in-a-cup” offers may persist, but most see these as ticket drivers, not loyalty builders. One technical root cause surfaced: “I see dirty burrs and lax descaling.” (Sean Dowell) A macro “wild card” was flagged (a temporary card-rate cap) that could nudge spend briefly.

Persona correlations and nuances

  • Older, rural/blue‑collar adjacent: Value predictable hours and facilities on highway travel; defect when taste is inconsistent and mobile orders clog throughput (Sean Dowell, Betty Crook, Allison Scavo).
  • Lower‑income, family‑oriented Hispanic: Price and community connection dominate; local panaderías feel better value and culturally affirming; bilingual/app frictions depress loyalty (Amanda Velasquez).
  • Midlife healthcare/service workers: Occasional users deterred by sugar‑heavy menu, unclear labeling, and religious/dietary constraints; halal uncertainty blocks food purchases (Linda Palomino).
  • Higher‑income frequent travelers: Still use Starbucks for travel convenience and mobile ordering, but defect to specialty shops for quality; skeptical of performative sustainability and rewards volatility (Valerie Guerra).
  • Cross‑income digital‑friction sensitive: App nags, tracking, complex rewards, and tip prompts reduce goodwill and app usage (multiple respondents).

Implications for an investable signal

Sentiment headwinds cluster around five themes: Taste, Price Pressure, Digital Friction, Operational Reliability, and Trust (sustainability, labor, ingredient transparency). Convenience provides a floor (airports, early hours, restrooms), but declining perceived value and execution risks weigh on advocacy and frequency. These themes are suitable for a composite sentiment index and event‑study analyses around pricing and LTOs.

Recommendations, risks, and measurement guardrails

  • Stand up a Starbucks Consumer Sentiment Index (SCSI) with five sub‑indices (Taste, Price Pressure, Digital Friction, Ops Reliability, Trust). Codify a taxonomy and seed from the 24 responses.
  • Expand data sources: App/Play Store and Google Reviews first; then Reddit/X, Google Trends, and news sentiment. De‑duplicate, geotag where possible, strip PII.
  • Backtest and monitor: Compute weekly lead/lag correlations to SBUX, directional hit rate, and event studies (pricing, LTOs).
  • Human‑in‑the‑loop labeling with bilingual gold sets to control drift and capture cultural/ingredient nuances (e.g., halal, sustainability).

KPIs: Coverage ≥5,000 mentions/week by Week 8; Pearson r of SCSI delta vs next‑week SBUX ≥0.20; directional hit rate ≥55% over rolling 26 weeks; macro F1 ≥0.80 across sub‑themes (EN+ES); median lead time ≥7 days to related news/earnings commentary.

Risks & guardrails: Sampling bias (mitigate via diversified sources and channel weighting), data/TOS/privacy (use official APIs, strip PII, audit logs), spurious correlation (hold‑outs, factor controls, confidence bands), taxonomy drift (active learning, periodic relabeling), pipeline reliability (SLIs/SLOs, retries, backfills).

Next steps

  1. Weeks 0–2: Finalize taxonomy; create Ditto collection/API; seed SCSI and a dashboard stub with current 24 responses and annotated quotes.
  2. Weeks 3–6: Go‑live SCSI MVP; add App/Play Store and Google Reviews; begin backtests; launch bilingual QA loop.
  3. Weeks 6–10: Add Reddit/X, Trends, and news; ship alerts on >1.5 SD moves; start directional hit tracking.
  4. Weeks 10–12: Pre‑earnings playbook; drift monitors and quarterly recalibration; share early correlation readout with confidence intervals.
  5. Governance: Complete legal/compliance review; enforce PII stripping and audit logging; document weighting and versioned taxonomies.
Recommended Follow-up Questions Updated Jan 11, 2026
  1. In the past 30 days, how many times did you get coffee or a similar beverage from each source? Provide a number for each, including zero: Starbucks; Dunkin'; McDonald's; convenience stores; local independent cafés; brew at home; other chain coffee shops.
    matrix Quantifies share-of-occasions vs competitors and home-brew to model traffic shifts and substitution risk tied to comp trends.
  2. On average over the past 30 days, how much did you spend per Starbucks visit (including tax and tip)? Enter a dollar amount.
    numeric Average ticket informs revenue per transaction and elasticity; pairs with frequency to estimate run-rate sales.
  3. What is the maximum price you consider acceptable for your usual Starbucks drink at your local store? Enter a dollar amount.
    numeric Price ceiling gauges elasticity headroom or risk from future price increases, informing margin and promo assumptions.
  4. Please rate Starbucks on the following attributes for your recent experiences (1=poor, 7=excellent): taste quality; value for money; speed of service; order accuracy; consistency across visits; mobile app ease; rewards program value; cleanliness; staff friendliness.
    matrix Structured diagnostics link specific operational levers to sentiment, enabling factor modeling of visit and spend changes.
  5. Which beverage or item category do you most often order at Starbucks?
    single select Category mix (cold beverages vs brewed coffee, food attach) links to margin and seasonality, improving revenue forecasts.
  6. How many Starbucks purchases do you expect to make in the next 30 days?
    numeric Near-term purchase intent provides a leading indicator for traffic trends within the panel.
Together these quantify traffic, ticket, mix, price tolerance, and operational drivers-key inputs to build a sentiment-to-sales index for correlation with SBUX performance.
Study Overview Updated Jan 11, 2026
Research goal: Track consumer sentiment toward Starbucks for correlation with stock performance; we asked about likelihood to recommend, change vs 6 months ago, last interaction, and 12‑month outlook.
Research group: Six US adults (ages 40–55) across rural/suburban and travel contexts with mixed incomes, including a lower‑income Hispanic parent, a healthcare worker, blue‑collar adjacent respondents, and a higher‑income frequent traveler.
What they said: Net endorsement is low (mostly 2–5/10), sentiment has nudged more negative, and outlook skews flat‑to‑slightly‑down despite continued, pragmatic use for convenience (ubiquity, early hours, clean restrooms, mobile).

Main insights: Drivers are consistent: perceived poor coffee taste (burnt/over‑roasted or too sweet), price/value mismatch, inconsistent execution, and growing digital friction (app nags, rewards changes, tip screens) that erode the convenience moat; staff are generally polite, so issues read as systemic.
Last interactions were functional but not enjoyable, and several cited trust/sustainability/labor or dietary‑labeling concerns (e.g., halal) that further depress loyalty; many are shifting routine coffee to home‑brew, c‑stores, or local cafés.
Most expect LTOs to support ticket in the near term, but without visible improvements to core taste, speed, and rewards simplicity, frequency is likely to drift down.
Takeaways:
  • Treat sentiment as a near‑term headwind: bias outlook to flat‑to‑down comps; watch for LTO spikes without loyalty.
  • Stand up a Starbucks Consumer Sentiment Index with sub‑indices (Taste, Price Pressure, Digital Friction, Ops Reliability, Trust) and track deltas vs next‑week SBUX returns.
  • Monitor leading signals: price/tip complaints, mobile‑order delays, app deletions, “thermos/home‑brew” mentions, and quality inconsistency.
  • Segment nuance: travel nodes remain resilient; Spanish‑speaking/price‑sensitive households and rural/blue‑collar cohorts are pulling back faster.