Maybe words like pilots, tests and experiments sound a bit risky right now. When time is tight, money is tighter and every decision feels like it matters, “experimenting” can sound like something large companies do with spare budgets and dedicated teams.
Maybe you are still getting the product ready, still sorting out your first listings, still figuring out whether your cousin’s enthusiastic “you should sell this” counts as market research or just family encouragement with snacks.
Fair enough. But the moment you start selling, you will start learning. A customer will ask a question you did not expect. A product photo will work better than another. A market day will bring compliments but not sales. A bundle will seem obvious to you and confusing to everyone else. A delivery promise will look manageable until life, load shedding, courier drama and packaging tape all form a committee.
That is when curiosity starts needing structure.
You change the price on Monday. Post a reel on Tuesday. Add a bundle on Wednesday. Join a market on Saturday. Update the packaging by Sunday night. Run a discount the next week because sales felt quiet. Then you change the caption, swap the product photo, add a delivery special, create a Mother’s Day offer, boost a post, open an online listing and ask three people whether the logo should be bigger.
Busy? Absolutely. Learning? Maybe not. And that is the interesting part. When you look at all that hustle, you realise you are already experimenting. The question is whether you are capturing what those experiments are teaching you.
This is where Amazon’s Leadership Principle “Learn and Be Curious” earns its work boots. Curiosity is not only reading more, asking questions or being open to ideas. In business, curiosity must eventually become disciplined learning. It must move from “let us try something” to “what did this teach us?” That difference matters.
Many South African sellers are not short of hustle. There is no shortage of movement. The country could run a backup generator on small-business effort alone. But effort without a learning trail becomes expensive noise.
The problem is not experimenting. Experimenting is healthy. A small business must test its way into better decisions because no founder can predict everything from the couch, no matter how majestic the couch or how strong the coffee. Customers surprise you. Markets shift. Costs move. Delivery partners have moods. The product everyone compliments may sell slowly, while the plain little item in the corner quietly pays rent.
The problem is changing too many things without knowing what changed the result.
If you reduce the price, change the photo, rewrite the caption, add free delivery and run a weekend special all at once, you may get more orders. Lovely. But what worked? Was it the price? The photo? The delivery offer? The timing? The fact that payday finally arrived wearing clean shoes?
You do not know. And if you do not know, you cannot repeat the win properly. That is why curiosity needs a notebook.
If you change five things at once and sales improve, you may feel clever, but you have not learned enough to repeat the win. Curiosity needs evidence, not just motion with a ring light.
Curiosity needs a notebook
If you have never “run an experiment” before, do not panic. You do not need a lab coat, a data analyst or a whiteboard covered in arrows. In seller language, an experiment is simply a small change made on purpose so you can learn something. One new bundle. One clearer product photo. One delivery option. One market day. One price test. One WhatsApp message to ten real customers.
The point is not to try everything. The point is to try one thing clearly enough that the result can teach you something.
In Lean and continuous improvement, PDSA is a simple learning cycle: Plan, Do, Study, Act. Plan the change. Do the change on a small scale. Study what happened. Act on what you learned. Keep it, change it, scale it or stop it.
For a seller, this does not need to become a formal improvement project with a spreadsheet that looks like it was raised by accountants. The practical version is an experiment receipt: one page, five questions, plain language, no theatre.
What are we trying?
Why are we trying it?
What do we expect to happen?
What actually happened?
What will we keep, change or stop?
That is it. That is the little receipt you keep after the business tries something.
Test one clear thing
Imagine a small candle business that wants to improve sales before the festive season. The founder could panic and change everything at once: lower the price, create gift boxes, add new scents, redesign the labels, offer free delivery, post daily and send a “please support my small business” message to every person who has ever made eye contact.
Tempting. Understandable. Slightly feral. But not very useful if the goal is to learn.
A better experiment might be smaller: test a two-candle gift bundle for two weeks with one clear photo, one clear price and one clear delivery promise. Before launching it, our would be millionaire writes down the expectation: “I think customers will choose the bundle because it feels easier for gifting and better value than buying one candle.”
Then the seller watches what happens. Did people ask about gift wrapping? Did they choose the bundle or still buy singles? Did the bundle increase order value? Did it create more packing work than expected? Did the margin still make sense after tissue paper, ribbon, label printing, courier costs and the emotional tax of trying to make a bow behave?
That is learning. The seller can now make a better decision. Keep the bundle. Change the price. Remove the ribbon. Add a gift note option. Offer it only during peak periods. Stop it completely because customers liked the idea but the profit left through the back door carrying a tiny suitcase.
This is the part many new sellers miss. Curiosity is not only asking, “What can I try?” It is asking, “What did I learn, and what will I do differently because of it?” Without that second question, the business becomes a hamster wheel with branding.
You see this often with pricing. Sales slow down, so the seller discounts. Sales increase, so the discount feels like proof. But if the seller never studies the result properly, the wrong lesson can sneak in.
The lesson may not be “customers only buy when it is cheaper.” It may be “the product sells better at month-end.” It may be “free delivery was more attractive than a lower item price.” It may be “the discount worked, but the margin was too thin to survive.” Those are different lessons. They lead to different decisions.
That is why a learning business must be careful with conclusions. A busy week does not always mean the experiment worked. A quiet week does not always mean it failed. Sometimes the product was good but the offer was not clear. Sometimes the seller accidentally ran the test during school holidays, tax season, a petrol price jump, three family birthdays and a national mood swing.
Real learning needs a little patience and a little honesty. The point is not to turn every small business into a laboratory where joy goes to fill in forms. The point is to stop paying for the same lesson twice.
If something works, write down why you think it worked. If something fails, write down what happened before you bury it in the backyard and pretend it was never your idea. If customers ask the same question during the experiment, capture it. If fulfilment becomes harder, capture it. If margin shrinks, capture it. If one channel performs better than another, capture it.
Your business is already teaching you. The question is whether you are taking notes.
Study the pattern, not your panic
This becomes especially important when a seller moves beyond friends, family and familiar buyers. Early feedback can be sweet, supportive and completely useless for decision-making. Your cousin may tell you the product is stunning because she loves you and wants peace at the next braai. That does not mean strangers will pay the same price, understand the offer or wait ten working days for delivery.
The market is less sentimental. Real buyers show you things that polite supporters may not say. They hesitate. They compare. They abandon carts. They write negative reviews. They ask about size, timing, ingredients, returns, quality and delivery. They buy the item you thought was boring. They ignore the product you personally loved. They choose the simpler bundle. They complain about a detail you assumed nobody noticed.
A curious seller does not take every signal personally. A curious seller studies the pattern. That is where PDSA protects both the business and the founder’s nervous system. Instead of treating every quiet week as rejection and every good week as destiny, the seller learns to ask better questions.
What did we change? What stayed the same? What did customers do? What did customers ask? What did the numbers show? What did the work feel like behind the scenes?
That last question matters. A promotion that creates sales but breaks the operation is not automatically a success. A market day that brings revenue but leaves the founder exhausted, underpaid and behind on orders may need a rethink. A corporate order that looks impressive but eats all the margin may be more ego snack than business meal.
Learning must include the customer, the money and the work. Otherwise, the seller may keep repeating tests that look good from the outside while quietly chewing the business from the inside.
Do not only test the shiny part
Let us make this practical.
If a spice seller wants to test a new braai bundle, the experiment should not only ask whether people liked the idea. It should ask whether customers understood what was included, whether the price made sense, whether the bundle increased order value, whether packing time stayed manageable and whether people came back for refills.
If a pet accessory seller wants to test a new harness size, the experiment should not only ask whether the colour looked cute. It should ask whether sizing questions reduced, whether returns decreased, whether customers sent fit feedback and whether the new size added useful sales or just created more stock complexity.
If a handmade skincare seller wants to test a refill pouch, the experiment should not only ask whether it feels eco-friendly. It should ask whether customers trusted the packaging, whether it leaked, whether it reduced cost, whether it changed repeat purchases and whether the refill message was clear enough for people to understand.
Three different sellers. Same lesson. Do not only test the shiny part. Test the business impact. This is where Learn and Be Curious becomes more mature than “I tried something.” It becomes a habit of closing the loop. Plan the experiment. Do it small. Study the result. Act on the learning. Then decide. Keep it because it worked and the operation can carry it. Change it because the idea was good but the execution needs adjustment. Scale it because the evidence is strong enough. Stop it because the result was weak, the cost was too high or the business has better places to spend its energy.
Stopping is also learning
Stopping is also learning.
That can be hard for founders because ideas have feelings attached to them. A product is not always just a product. It can be a late night, a brave decision, a personal taste, a hope, a little piece of identity wearing a price tag. So when an experiment does not work, it can feel like the market is insulting your soul in public.
But a failed test is not always a failed business. The gift bundle did not sell. Good. Now you know. The new caption did not improve enquiries. Fine. Now you know. The market stall brought compliments but no profit. Useful. Now you know. The discounted price brought sales but damaged margin. Painful, but now you really know.
That knowledge is valuable if you use it.
The danger is pretending to learn while actually collecting random activity. This is how sellers end up exhausted, confused and surrounded by half-finished ideas: a newsletter they sent twice, a TikTok account they resent, packaging they cannot afford, a bundle nobody asked for, a special that trained customers to wait for discounts and a drawer full of branded stickers that may one day need their own support group.
Write down the test. Give it a time frame. Keep the change small enough to understand. Watch the result. Make a decision. That is not overcomplication. That is respect for your own effort. Because effort is expensive. Time is expensive. Stock is expensive. Hope is expensive when it keeps replacing evidence.
Learn and Be Curious asks the seller to stay open, but not scattered. It asks the seller to try things, but not blindly. It asks the seller to listen to the market, but not chase every noise like a dog after a Hadeda.
The goal is not to become perfect at experimentation. The goal is to stop letting useful lessons disappear into memory, panic or vibes.
Curiosity starts the experiment. Discipline turns it into learning. And that is how a seller stops guessing in circles and starts building with evidence.
This is a personal thought piece, written from my own customer experience and process improvement perspective. It draws on publicly available information and reflects my own views.