How to understand your customers at the farmers market

# You are running blind, and the format is why

Direct-to-customer selling is not a fringe channel. The USDA’s 2022 Census of Agriculture counted more than 116,000 farms selling food straight to the people who eat it, some $3.3 billion worth, and its directory has listed close to 8,800 farmers markets nationwide. The pull for shoppers is well documented: a 2021 study in PLOS ONE found freshness is the single most important thing market shoppers look for, ahead of price, and US research has long found people choose markets to support local growers and to buy straight from the person who made the food. The irony is that the format built on a personal relationship gives the seller almost no way to remember it.

Think about what a single sale tells you. Online, it is a row of data: who, what, when, where they came from, what they nearly bought instead. At your table it is a coin and a smile. You cannot tell a first-timer from a tenth-time regular, you cannot count how many people considered the jam and walked on, and you cannot connect the slow Saturday to anything you could actually change. The relationship is the whole point of a market, and it is the one thing the transaction never records.

# The customers you most need to hear from already walked off

The feedback you do get is wildly skewed. The person who loved the peaches tells you they are the best of the summer. Everyone else just leaves. The shopper who thought your pint was overpriced does not argue the point, they drift to the next stall. The one who found the line too long, or could not tell what the green things were, or assumed you only took cash, says nothing at all. Most unhappy customers never complain. They simply do not come back, and you never learn why.

That silent majority is the most valuable audience at the market, and the hardest to reach. It is easy to believe you are doing great when the only people who speak up are the ones already sold. Bain & Company put a famous number on that blind spot in 2005: 80% of companies believed they delivered a superior experience, while only 8% of their customers agreed. A market stall is that gap in miniature, played out a few hundred times every Saturday, mostly in silence.

# The catch-22: the busiest stall learns the least

Here is the trap every vendor knows in their body. The moment you could ask a customer what they think is exactly the moment you have no time to do it. During the rush you are weighing, bagging, making change, and keeping an eye on the cash box, often single-handed. You cannot run a survey with your hands full. Then the crowd thins, you finally have a minute, and there is no one left to ask. The window when asking is possible and the window when it matters never overlap.

This is why the usual advice to “talk to your customers” quietly fails at a market. You are not avoiding the conversation, you physically cannot have it at the only time it would be representative. So the asking has to happen without you: something that invites every passerby to weigh in on their own, while you get on with selling, and hands you what they said once the day is done.

# Let the stall do the asking

The cheapest version of this is a chalkboard. Write one question on the A-frame by your table, “What should we bring more of next week?”, and leave room for people to think about it. It costs nothing and it beats asking no one. But a chalkboard cannot capture an answer, so you are back to overhearing the few who say something out loud.

The upgrade is to make answering self-serve and to capture every reply. Prop a small screen or a tablet on the table, or pin a printed QR code to the stall, showing one question and an invitation to scan. People answer on their own phone, in their own words, in the few seconds they are standing in front of you deciding. No app to install, no sign-up, no line. It runs the whole market without your attention, which is the only kind of feedback collection that survives a Saturday rush.

And the gear costs almost nothing. A basic Android tablet runs about $40, and since a market stall rarely has Wi-Fi, you just connect it to your phone for an internet connection during the hours you are open. That is the entire setup: a cheap tablet propped on the table, sharing your phone’s signal, asking the question all morning while you sell.

# Ask one question anyone answers in ten seconds

Whether you use a board, a card, or a screen, the question makes or breaks the response rate. Ask one thing, in plain words, answerable in a sentence from a phone someone is already holding. A long form with ratings and boxes gets ignored at a market even more than it does online, because nobody stops their Saturday to fill one in. Keep it open-ended rather than multiple choice: a thumbs-up tells you the mood, but the words tell you why, and the why is what you can act on next week.

Pick the question to match what you actually need to decide. If you are choosing what to grow or make, ask about that. If sales are soft, ask the walk-by crowd what held them back. And keep asking the same thing for a few weeks so you can compare like with like, rather than rewording it every market and losing the thread.

For the people who bought

What made you stop at our table today? · What should we bring more of next week? · What is one thing you wish we sold?

For the walk-by crowd

What almost stopped you from buying today? · Was anything unclear, the prices, the products, how to pay? · If you ran this stall, what would you change first?

# Capture the “no,” not just the “yes”

The single most useful thing you can do is place the question where the people who did not buy will still see it. A board angled toward the aisle, a QR code on the edge of the table or on the back of your business card, a screen they pass whether or not they stop. The buyers will tell you plenty on their own. It is the near-misses, the ones who slowed down and moved on, whose reasons you have never once heard.

Anonymous is the right default here. People are far more honest about why they did not buy when there is no name attached and no salesperson watching them type. You are not after a clean headcount, you are after the shape of what people think, and a scan-and-type with no sign-up gets you the most candid version of it.

# A hundred replies, one clear read

Collecting answers is the easy half. Reading them is where most feedback efforts die. A busy market can hand you a hundred short replies in a morning, and no vendor is going to sit down Saturday night and read a hundred comments, so the pile goes unread and the whole exercise quietly stops. The richest thing you have becomes the thing you ignore.

This is the part worth getting help with. Instead of a hundred separate notes, you want one short, faithful summary of what people said today: the recurring request, the common gripe, the thing people kept praising, in a paragraph you can read between customers. That is exactly what One Voicer does with the replies to a question. It reads every voice and blends them into a single clear answer that updates as more arrive, so by the time you are packing up, the read on the day is already done.

# Track the season, not just Saturday

A single market tells you about a single day, and a single day at a market is mostly noise. Sales swing on the weather, a holiday weekend, a competing market across town, who else showed up. To learn anything you need each week saved so you can read the trend instead of the last data point. Save a weekly snapshot of the blended answer and over a season you build a dated history: a short, distilled read for each market, stacked so you can see what is actually changing.

That is the difference between guessing and knowing. “Sold out” might mean a product is your most-wanted, or it might mean you under-stocked it, and only a run of weeks tells you which. What you measure consistently, you can manage. The idea is old: Frederick Reichheld argued in Harvard Business Review back in 2003 that one number, tracked faithfully over time, beats a binder of one-off survey results. A season of weekly summaries is that kind of measurement, in a form you can actually read.

# Close the loop so they come back

The last step is the one most vendors skip, and it is the cheapest loyalty you will ever buy. When people see that you listened, they come back to see what else changed. So tell them. A line on the chalkboard next week, “You asked for more heirlooms, they are back by the scale,” does more for repeat business than any punch card, especially when you have no email list to fall back on.

It does not have to be big. One thing people raised, one thing you changed, shown where the crowd can see it. Feedback that visibly goes somewhere keeps coming, and keeps getting more honest. Feedback that vanishes into a box trains people to stop bothering, which is how most stalls end up flying blind in the first place.