Food Delivery Survey Template
Run a fast, mobile-friendly food delivery survey right after drop-off to track satisfaction and pinpoint what drove it (ETA, accuracy, food condition, packaging, courier, and support). Choose a short 5-question pulse for most orders or a 10-12 question diagnostic version when you need root causes. Use the trigger rules and routing steps below to turn feedback into fixes by zone, daypart, and partner.
Trusted by 5000+ Brands
Who to Survey (and When): Post-Dropoff vs Post-Issue Triggers
Goal: Capture accurate feedback while the delivery details are still fresh.
Set it up: Trigger a post-dropoff survey 15-60 minutes after the "delivered" event (and no later than 48 hours).
Example: If the order arrived with cold food, send the 5-question pulse by SMS 30 minutes after drop-off.
Send windows that work in delivery
- Standard delivered orders: Send 15-60 minutes post-dropoff for the best recall on ETA, item accuracy, and food condition.
- Longer comments (optional): If you want richer verbatims, add an email follow-up the next morning (but keep it sampled).
- Issue-triggered orders (late/missing/incorrect): Send after the case is resolved so customers can rate the fix, not the chaos.
First-time vs repeat customers (fatigue control)
- First-time customers: Survey more often for the first 1-2 deliveries to catch onboarding and expectation gaps.
- Repeat customers: Cap invitations (for example, 1 invite per customer every 14-30 days), unless an order is flagged as an issue.
- High-frequency buyers: Rotate who you sample so you do not over-hear from the same people.
Delivered event: If status = delivered, then send CSAT pulse in 15-60 minutes.
Delay flag: If ETA miss > X minutes (pick 10-20), then add the "late" follow-up question.
Refund/credit closed: If case status = closed, then send CES (effort) within 1-4 hours.
Data rule: Keep identifiers minimal and avoid sensitive fields in-survey. Use privacy and data minimization as your default: make order ID optional, and do not ask for full address or payment details.
CSAT vs CES vs NPS for Food Delivery (Which to Use, When, and Why)
| Metric | Best for | When to ask | Where it fits in the flow | Common pitfall | What you decide with it |
|---|---|---|---|---|---|
| CSAT (order satisfaction) | How good this specific delivery was (ETA, accuracy, condition). | Every delivered order (or a rotating sample), 15-60 minutes post-dropoff. | First question in your post-delivery survey. | Mixing timeframes ("this order" vs "overall"). | Which ops drivers to fix first (late, missing items, packaging failures). |
| CES (effort to fix a problem) | How hard it was to get help (refund/credit, re-delivery, support). | Only when support was used or an issue was filed. Ask after resolution closes. | Issue path or separate post-case survey. | Asking everyone CES even when nothing went wrong. | Where customers get stuck (contact, handoff, policy, speed of refund). |
| NPS (loyalty) | Relationship signal that moves slower than one order. | Monthly/quarterly, or after 2-3 deliveries (sampled). | Separate program or appended to a periodic check-in. | Asking NPS after every order and reacting to single-order noise. | If you are building repeat behavior (retention, subscription, referral). |
Decision rule: Keep CSAT as your always-on post-delivery metric. Add CES only when a case closes. Run NPS on its own cadence, not every receipt.
- If you need fast ops fixes: CSAT + driver questions by zone/daypart.
- If support cost is the problem: CES after refunds/credits close, plus one "what was hard" follow-up.
- If leadership wants a loyalty KPI: NPS monthly/quarterly, and read it as a trend line.
Different loyalty and satisfaction metrics do not behave the same, so do not expect one "best" score across use cases. Use each metric for its decision and watch the trend over time (not a universal target), consistent with findings in The value of different customer satisfaction and loyalty metrics in predicting business performance.
Question Sets by Delivery Journey Stage (5-Question Pulse + 10-12 Standard)
Goal: Diagnose what drove satisfaction for one specific order.
Set it up: Default to a 5-question pulse on mobile, then expand to 10-12 questions when CSAT drops.
Example: If CSAT falls in one zone at dinner, add packaging and courier questions for that zone only.
5-question pulse (keep it on almost every delivered order)
"Overall, how satisfied were you with this delivery?"
Why it matters: This is your top-line order CSAT. You will trend it by zone, daypart, and partner.
When to use: Ask on every post-dropoff survey. Keep the timeframe as "this delivery".
"Was your order delivered within the estimated time?"
Why it matters: ETA misses are a common reason CSAT drops, even when food is correct.
When to use: Ask on every order. If "No", show a follow-up for minutes late (0-10 / 11-20 / 21-40 / 40+).
"Were any items missing or incorrect?"
Why it matters: Accuracy failures create refunds, rework, and repeat contacts.
When to use: Ask on every order. If "Yes", show an item-level prompt (missing / wrong item / wrong customization).
"How was the food when it arrived (temperature and condition)?"
Why it matters: This separates "delivery was fast" from "food traveled well".
When to use: Ask on every order using a short Likert scale (for example, 1-5 from "Very poor" to "Excellent").
"What is the one thing we should change to improve deliveries like this?"
Why it matters: You get a fix, not a rant. This drives your weekly ops backlog.
When to use: Ask last. If CSAT is low (for example, 1-2 out of 5), show a second box: "What went wrong?"
10-12 question standard (turn on when you need root causes)
"How accurate was the estimated delivery time shown at checkout?"
Why it matters: A "late" experience often starts with an over-promised ETA.
When to use: Add when late delivery is a top driver. Use a 1-5 rating and trend by zone and daypart.
"How would you rate the packaging (sealed, spill-resistant, and organized)?"
Why it matters: Packaging failures show up as spills, sogginess, and tamper concerns.
When to use: Add when you see drops in food condition or complaints about mess. Route issues to the packaging owner.
"How professional was the courier (communication and handling)?"
Why it matters: Courier behavior can explain low scores even when kitchen performance is strong.
When to use: Add if you manage courier partners or see patterns by partner ID. Keep wording behavior-based.
"Were delivery fees and totals clear before you placed the order?"
Why it matters: Price surprises can drag down satisfaction even when delivery execution is fine.
When to use: Add if you are testing fee changes or see low scores tied to specific platforms.
"Did you contact support about this order?"
Why it matters: This is your fork in the road for CES and recovery questions.
When to use: Ask on all standard surveys. If "Yes", show CES and one recovery outcome question.
"How easy was it to get your issue resolved?"
Why it matters: This is CES. It tells you how much work customers do to get a refund/credit or replacement.
When to use: Show only if support was used or an issue was filed. Ask after the case closes when possible.
Watch for driver patterns that repeat across studies in delivery quality (timeliness, accuracy, and food condition often move satisfaction and reuse). Use the standard version when you need those drivers separated clearly, as modeled in Modelling the significance of food delivery service quality on customer satisfaction and reuse intention.
Distribution, Sampling, and Fatigue Control (SMS, Email, In-App)
Goal: Get enough feedback per zone and daypart without burning out repeat buyers.
Set it up: Start with a short census for 1-2 weeks, then switch to sampling once you know response rates.
Example: If dinner volume is high, sample 20%-40% of delivered orders and rotate customers weekly.
Pick a channel based on speed vs depth
SMS: Best for 15-60 minute post-dropoff pulses. Keep it to 5 questions.
Email: Best when you want longer comments. Use it for sampled diagnostics, not every order.
In-app: Best for logged-in users and lower cost. Trigger on the delivered screen or next app open.
Decide: census vs sample
Run a census when volume is low, or when you are launching a new zone or courier partner.
Switch to sampling when repeat buyers start seeing too many invites. Set a customer cap (for example, 1 invite per 14-30 days).
- Stratify: Sample by zone, daypart, and platform so one segment does not dominate.
- Rotate: If you sample 30% today, rotate who is eligible tomorrow.
- Plan targets: Use sample size guidance to set minimum completes per zone before you trust a pattern.
Reduce bias and keep results believable
Watch for response bias, where extreme experiences answer more often than average orders.
Set one light reminder (for example, 24 hours later) and stop after that. Then track response rates by zone, daypart, and platform.
If one segment responds less, increase sampling there instead of blasting everyone.
Use representativeness checks like the ones described in Assessing the Representativeness of Public Opinion Surveys.
How to Turn Food Delivery Feedback into Fixes (Routing + Dashboard)
Goal: Turn low CSAT and angry comments into specific fixes (prep time, picking, packaging, courier handling, refunds).
Set it up: Review results weekly, route issues daily, and hold owners to simple SLAs.
Example: If CSAT drops in one zone, check ETA accuracy first, then packaging and courier ratings.
- Step 1: Validate the top-line by the way you operateCut CSAT (and CES/NPS if you run them) by zone/location, daypart, platform, courier partner, and order type. Watch for sudden drops and slow drifts. Set a minimum completes rule per segment before you act.
- Step 2: Tag open-text into driver buckets you can assignUse 6-8 tags: ETA, missing/incorrect items, food temp/quality, packaging, courier professionalism, app/checkout, fees, support. Keep tagging lightweight so you can do it every week.
- Step 3: Prioritize with frequency x impactFrequency = how often the tag appears. Impact = CSAT (or CES) delta when the tag appears. Fix high-frequency, high-impact items first (for example, missing sides at dinner or spills on soups).
- Step 4: Route issues to owners with clear SLAsRoute by tag: kitchen for accuracy, ops for prep timing, courier/partner manager for handling, support for refunds. Set simple SLAs (for example, acknowledge in 1 business day, resolve in 3). Align routing and recovery steps with practices in ISO 10002 complaints handling guidelines.
- Step 5: Close the loop and prove the fix workedClose the loop on the highest-friction cases (late by 40+ minutes, missing core items, spilled orders). Then re-check CSAT and driver ratings in the affected segment. Use a simple monitoring cadence consistent with ISO 10004 monitoring and measuring customer satisfaction guidance.
Dashboard outline (keep it simple)
- Top line: CSAT trend (weekly) and CES trend (issue-only).
- Drivers: ETA met %, missing/incorrect %, food condition rating, packaging rating, courier rating.
- Cuts: Zone/location, daypart, platform, courier partner, restaurant/ghost kitchen.
- Action queue: Top 10 themes by frequency x impact, with an owner and due date.
Frequently Asked Questions
When should I send a food delivery satisfaction survey?
Send your post-dropoff survey 15-60 minutes after delivery so customers can recall ETA, missing items, and food condition. If you want longer comments, add a sampled email the next morning. For late/missing/refund cases, trigger the survey after the issue is resolved so you can measure effort and recovery.
Should I use CSAT, NPS, or CES for delivery orders?
Use CSAT for this specific order, right after drop-off, to find operational drivers like ETA misses and accuracy errors. Use CES only when support or an issue flow happened, and ask it after the refund/credit closes. Use NPS on a separate monthly or quarterly cadence (or after 2-3 deliveries) to track overall loyalty without mixing timeframes.
How long should a food delivery survey be on mobile?
Keep your default mobile survey to a 5-question pulse so completion stays high. Use a 10-12 question version only when you need diagnostics, and show extra questions only when relevant (late delivery, missing items, or support contact). Avoid long grids and keep most questions to single taps.
How do I avoid biased results (only angry or happy customers responding)?
Use consistent triggers (delivered event and issue flags) and rotate sampling so you do not only hear from extremes. Set one light reminder, then stop, and monitor response rates by zone, daypart, and platform. Focus on trends and segment deltas instead of chasing one universal benchmark.
What customer data should I (and should I not) collect in the survey?
Collect only what you need to diagnose the order: optional order ID, zone/location, platform, and first-time vs repeat. Avoid sensitive data like full address, payment details, or anything you can already join from your order system. Keep identifiers minimal and store them outside the survey when possible.
What should I do with open-ended comments from delivery surveys?
Tag comments into a small set of buckets (ETA, accuracy, food temp/quality, packaging, courier, app/fees, support) so you can count them weekly. Prioritize fixes by frequency x impact using CSAT/CES deltas, then route each theme to an owner with an SLA. Close the loop on the biggest failures and check if the driver scores improve in the affected zone or daypart.
Related Survey Templates
FREE TO START -- NO CREDIT CARD REQUIRED