How to Test Your Business Idea Before Spending Your Capital: 7 Tests for Under EGP 5,000

Most ideas do not fail because their owners did not work hard. They fail because the owner spent the capital setting the business up before confirming that anyone would buy. The good news is that testing demand is far cheaper than most people think. This article sets out seven practical tests, ordered from free to most expensive, that can be run in Egypt for under EGP 5,000, with how to read each result and how to carry it into a feasibility study.

Why test before you spend?

In CB Insights' latest analysis of why venture-backed startups have shut down since 2023, running out of capital was cited in 70% of cases and poor product-market fit in 43%. The report itself notes that running out of money is usually the final cause of death rather than the root problem¹. The two are linked: a business selling something not enough people need burns its money waiting for sales that never come.

In Tom Eisenmann's research on why startups fail, the Harvard Business School professor of entrepreneurship identified a prominent pattern he called the "false start": a founder launches before researching customer needs thoroughly enough, then discovers they have built the wrong thing after spending much of their money and time².

The core idea is old and simple: every assumption in your plan is a guess until you test it. Eric Ries, who developed the Lean Startup method, defines the "minimum viable product" as the version of a new product that lets you collect the maximum amount of validated learning about customers with the least effort³. The goal is not to build a small version of the whole business but to find the cheapest way to answer one question: will people pay?

Before you start: write down your assumptions

Each test answers a specific assumption. So before anything else, write down the assumptions your idea rests on and rank them by risk, meaning how much damage it would do if they were wrong. In most small businesses the riskiest assumptions are:

1. The problem: do enough people have the problem you solve, and do they care about solving it?
2. The customer: who exactly is the customer, and where do you find them?
3. The price: will they pay a price that gives you a sufficient margin?
4. Reach: how much will it cost you to reach one customer who actually buys?
5. Repeat purchase: will they buy again, or only once?

Then, for each test, set the success and failure criteria before running it. For example: "If at least 10 out of 200 page visitors order, we move to the next step. If fewer than 3 do, we revisit the idea or the price." These numbers are not fixed scientific rules but limits you set in advance, so you do not persuade yourself after the result appears that a weak figure is "encouraging".

Test 1: 20 interviews with potential customers (free)

The cheapest and most useful test is to talk to 20 people from the customer group you are targeting. Not your friends or family, but people who fit the description of the customer you expect: mothers in a particular area, workshop owners, students at a particular university.

The most important rule in the interview: ask about the past, not the future. "Would you buy this product?" gets polite answers that are worthless. Useful questions look like this:

• When did you last face this problem? What did you do?
• How much did you pay to solve it? Where did you buy?
• What bothered you about the solution you used?
• Did you look for an alternative? Where did you look?

How to read the result: if most of the people you spoke to cannot remember the last time they faced the problem, or have spent no money or time on it, the problem is probably not painful enough to build a business on. If they describe workarounds they use, or have paid for alternatives they dislike, that is a strong sign of a real need.

Test 2: study competitors from your desk (free)

Having competitors is not bad news; it is often evidence that demand exists. The questions are what they offer, at what price, and what their customers complain about. You can learn a lot without spending a pound:

• Google Maps: search for the type of business in your target area, count the competitors and read the negative reviews in particular. Recurring complaints are your opportunity.
• Meta Ad Library: lets you search all active ads running across Meta's platforms, including Facebook and Instagram⁴. Look up your competitors' pages and note the offers, prices and messages they repeat. An ad that keeps running for weeks is usually one that works.
• Social pages and online stores: review prices, customer comments, recurring questions and response times.
• Visit as a customer: if the competitor has a shop, buy from it once. One purchase costs far less than what you will learn about the service and the customer traffic.

How to read the result: make a table of competitors' prices, strengths and weaknesses. If you find no competitors at all, ask yourself why: is the idea genuinely new, or did those who tried it before you fail? If you find many competitors with similar prices, you will need a clear advantage or a lower cost.

Test 3: are people searching for a solution? (free)

Google Trends shows relative interest in a search term over time and by region. on a scale of 0 to 100 calculated from a sample of searches, so it does not show actual search counts⁵. That makes it useful for comparison, not calculation: is interest in a term such as "home nursery" rising or falling? Is it seasonal? Which governorates is it concentrated in?

Also try typing search terms into Google and YouTube and look at the autocomplete suggestions: they reflect what people actually search for. And search local Facebook groups for people's questions about the problem you solve.

How to read the result: steady or rising interest is a good sign. Sharp seasonal interest means your feasibility study must allow for weak months. No searches at all may mean people do not know a solution exists, which raises marketing costs, because you will need to make them aware of the problem before you can sell them the solution.

Test 4: an offer page with pre-orders (EGP 500–1,000)

After the free tests, it is time to present the product as if it exists and see who orders it. Create a simple page, on a free website builder, a social media page or an order form, that clearly explains the product or service, the price, the expected delivery date, and an order or booking button.

The product does not need to be ready. You just need to be honest: "We launch next month; reserve your place now." If you have no product photos, use a prototype or an illustration and say clearly that it is a concept.

What to measure: the number of visitors, the number who clicked the order button and the number who completed their details. The share of visitors who complete an order is the "conversion rate", one of the most important figures your feasibility study will need later.

Test 5: a small paid ad (EGP 1,000–2,000)

A page is not enough if nobody visits it. The quickest way to bring visitors from your target group is a small paid ad on Facebook and Instagram, platforms with a wide reach in Egypt: DataReportal estimates there were about 98.2 million internet users in Egypt at the end of 2025 and about 51.6 million social media user identities, which are accounts rather than necessarily distinct people⁶.

Reaching people online does not mean they buy online, though. The U.S. International Trade Administration estimates that only 8.3% of Egyptians make regular online purchases⁷. So do not make advance online payment the only option in your test; also allow ordering through WhatsApp or cash on delivery, and measure each separately, because the method your customers prefer will affect collection and delivery costs in your study.

Write two or three ads with different angles: one focused on the problem, one on price and one on speed or quality. Split the budget evenly over several days, then compare:

• Cost per click: how much did you pay for each visitor who reached the page?
• Cost per order: how much did you pay for each person who completed an order or booking? This is the closest figure to the customer acquisition cost in your study.
• Which angle worked: the message that produced the cheapest order is usually what matters most to your customers.

How to read the result: compare the cost per order with the margin from a single customer. If the contribution margin on the first purchase is EGP 100 and acquiring the customer costs EGP 300, the business only makes money if the customer buys more than three times. That turns the question into: is your product something people buy repeatedly?

Test 6: ask for a deposit (almost free)

Clicking "order" is evidence of interest, but paying is evidence of demand. So the strongest test is to ask those who showed interest for a small deposit or advance payment, with a clear promise to refund it if the product does not launch on time.

This has become easier in Egypt with the spread of instant payments: according to the Central Bank governor, the Instant Payment Network (InstaPay) handled about 1.47 billion transactions in the second quarter of 2026 and had about 22 million registered customers⁸, so you no longer need a payment gateway or merchant account to collect a small deposit. Record every amount precisely, keep a clear list of who paid, and honour refunds if you do not launch.

How to read the result: compare the number who paid with the number who ordered. A large share of orders without payment means the interest is shallow, or the price is higher than customers will accept when it is time to pay. On the other hand, 10 people paying a deposit for a product not yet made is stronger evidence than 100 likes.

Test 7: deliver by hand on a small scale (EGP 1,000–2,000)

Before renting a shop or buying equipment, deliver the product or service yourself to a few customers, even if the method is inefficient. This is the minimum viable product in its simplest form³:

• Restaurant or food product: cook from a licensed kitchen or one rented by the hour, sell a limited number of orders in a week, or take part in an exhibition or a Friday market.
• Service: deliver it to five or ten customers at its real price, do it yourself and record the time each one takes.
• Online store: buy a small quantity from the supplier, or sell on pre-order, before buying large stock.
• App or platform: deliver the service by hand through WhatsApp and spreadsheets before writing any code.

What to measure: the actual unit cost (not the estimate), the time delivery takes, customer satisfaction and, most importantly, whether anyone came back to buy again without being asked.

Three examples: choosing the right tests for your idea

Not every business needs all seven tests in the same order. The examples below show how to choose what suits your idea. They illustrate a way of thinking; they are not real results from specific businesses.

A home bakery for healthy sweets

The riskiest assumption here is price: healthy sweets cost more, so will customers pay the difference? Start with interviews to learn who buys these products now and where, then study competitors' pages and prices. Next, show a short menu of three products on a simple page, with an ad targeted at a defined delivery area, and ask for payment on order or a deposit for large orders. During the hands-on trial, calculate precisely the cost of ingredients, packaging and delivery per box, because those numbers will set your real margin.

A training centre or classes in a specific skill

The riskiest assumption here is reach: can you fill a full group at a reasonable marketing cost? The search-interest test reveals seasonality, since demand for some courses rises in the holidays and drops at exam time. Then advertise a first group with a fixed date and ask people to reserve a place with a deposit. If the group fills, run the first course in a hall rented by the hour instead of leasing permanent premises. The share who sign up for the next course is the most important number you will come away with.

An online store for imported products

The two riskiest assumptions here are customer acquisition cost and the capital tied up in stock. Study competitors' ads in the Meta Ad Library to see which products they promote consistently, then offer products on pre-order before importing a large quantity. The small ad tells you what one order costs you, and the pre-orders tell you which products people actually order. That way your first shipment is bought on real demand, not on a forecast.

Budget for the seven tests

The figures in the table are budgets you set, not market prices. You can easily stay under EGP 5,000 by starting at the lower end for the ad and the hands-on trial, then increasing the budget for whichever test gives promising results. Compare that sum with what you could lose by fitting out a shop or buying equipment for an untested idea.

How to read the results and decide

After the tests, you will face one of three outcomes:

1. Strong signals: people describe the problem clearly, order, pay deposits, and the cost of acquiring a customer is below the margin you expect from them. Move on to a detailed feasibility study built on your real numbers.
2. Mixed signals: clear interest but weak payment, or a high acquisition cost. Do not abandon the idea; change one variable and test again: the price, the target group, the message or the sales channel.
3. Weak signals: interviews reveal no painful problem, and nobody orders even at a low price. You have lost a few thousand pounds instead of all your capital, and that in itself is a good result.

Avoid two common mistakes when reading results. The first is changing more than one variable at a time, so you cannot tell what improved or weakened the result. The second is relying on the opinions of people close to you, who tend to be kind and do not represent the customer who will pay.

Common mistakes when testing an idea

1. Asking about intentions rather than behaviour: "Would you buy?" gets a polite answer; "What did you buy last time?" gets a fact.
2. Testing a price below the real one: if you show a discounted price to attract orders, you are testing a different idea from your business. Show the price you will actually charge.
3. Spending the whole budget on one test: spread it, start small, and spend more on what produces results.
4. Advertising to too broad an audience: an ad aimed at "all of Egypt" gives you numbers you cannot interpret. Target the area and group you will actually serve.
5. Ignoring repeat purchases: a first sale is not enough for most businesses. Watch who comes back, because a customer who buys once may not cover the cost of acquiring them.
6. Testing forever: set a period for each test, a week or two, then decide. A test is a way to make a decision, not a substitute for one.

From testing to a feasibility study

The real value of the tests is that they turn the study's assumptions into figures from the market. Instead of writing "we expect 5% of visitors to buy", you write "in a test with 400 visitors, 4% bought". Instead of "marketing costs 10% of sales", you write "customer acquisition cost in the test was EGP 120". These figures also matter to whoever reads the study: a partner, an investor or a bank.

This is where Kleverz's two tools fit in: the free idea check helps you organise your assumptions and identify the riskiest before you start testing, and the detailed study turns your test results into a sales forecast, a monthly cash flow and a break-even point. The more the figures you enter in the form come from a real test, the closer the study is to reality. We explain how to calculate break-even on these figures in our guide to break-even and working capital.

Frequently asked questions

Sources

1. CB Insights. The top 9 reasons startups fail (2026-03-05). Checked 14 October 2026.
2. Harvard Business Review. Eisenmann, T. Why Start-ups Fail (May–June 2021) (2021-05). Checked 14 October 2026.
3. Lean Startup Co.. Ries, E. What is an MVP?. Checked 14 October 2026.
4. Meta Transparency Center. Meta Ad Library tools (2023-08-24). Checked 14 October 2026.
5. Google Trends Help. FAQ about Google Trends data. Checked 14 October 2026.
6. DataReportal (Kepios). Digital 2026: Egypt (2025-11-08). Checked 14 October 2026.
7. U.S. International Trade Administration. Egypt – eCommerce (Country Commercial Guide) (2025-11-21). Checked 14 October 2026.
8. Ahram Online. Egypt's IPN records 1.47 bln transactions worth EGP 4.33 tln in 2Q 2026: CBE governor (2026-10-01) — secondary source. Checked 14 October 2026.