There is a particular kind of expensive lesson that most growth-stage businesses eventually learn.<\/p>\n\n\n\n
The strategy felt obviously right in the planning meeting. Everyone agreed. The campaign was approved. The budget was committed. The team started executing.<\/p>\n\n\n\n
Three months later, the result is indistinguishable from doing nothing.<\/p>\n\n\n\n
Then comes the difficult question:<\/p>\n\n\n\n
Why did we not know sooner?<\/strong><\/p>\n\n\n\n The answer is often uncomfortable.<\/p>\n\n\n\n The strategy was never tested properly before the business committed to it at scale.<\/p>\n\n\n\n There is a tendency to treat validation as hesitation.<\/p>\n\n\n\n The thinking goes something like this: confident businesses make decisions and move quickly. They do not spend weeks testing an idea before committing.<\/p>\n\n\n\n That gets the economics backwards.<\/p>\n\n\n\n A small experiment is not a hedge against confidence. It is the cheapest form of information your business can buy.<\/p>\n\n\n\n The alternative is buying that information later, at full campaign scale, after a significant portion of the quarter’s budget has already been spent.<\/p>\n\n\n\n By then, the business has fewer options.<\/p>\n\n\n\n The budget is committed. The team is emotionally invested. The campaign may already have created operational commitments. And stopping can feel like admitting failure.<\/p>\n\n\n\n That is exactly why validation needs to happen before the stakes become large.<\/p>\n\n\n\n The strongest businesses are not necessarily the ones with the boldest convictions. They are the ones that have made being wrong at small scale a normal part of how they operate.<\/p>\n\n\n\n Be wrong cheaply. Learn quickly. Scale what survives.<\/strong><\/p>\n\n\n\n Validation is the process of testing an important assumption before making a larger investment based on that assumption.<\/p>\n\n\n\n The assumption could relate to:<\/p>\n\n\n\n The purpose is not to prove that your idea is brilliant.<\/p>\n\n\n\n The purpose is to discover whether your idea deserves more investment.<\/p>\n\n\n\n That distinction changes how the experiment is designed.<\/p>\n\n\n\n Validation starts with a good assumption, and a good assumption usually comes from good diagnosis.<\/p>\n\n\n\n Before you test a campaign, product or offer, you should understand what you are actually trying to prove.<\/p>\n\n\n\n That is why the Growth Lab<\/a> begins with understanding the customer, the business and the market before major decisions are made.<\/p>\n\n\n\n Validation should also sit inside a measurable commercial framework. A test should have a clear objective, a defined audience, an appropriate channel and a measurable outcome.<\/p>\n\n\n\n The sequence is important:<\/p>\n\n\n\n Diagnose the problem \u2192 define the position \u2192 engineer the approach \u2192 validate the assumption \u2192 scale the result.<\/strong><\/p>\n\n\n\n An experiment only produces useful information if failure is genuinely possible.<\/p>\n\n\n\n This sounds obvious, but many SME tests quietly avoid the possibility of a negative result.<\/p>\n\n\n\n The budget is too small to produce a useful signal. The test runs for too little time. The target audience is too broad. The success metric is unclear.<\/p>\n\n\n\n Or, most commonly, nobody agrees beforehand on what \u201cthis didn’t work\u201d actually means.<\/p>\n\n\n\n Then the result arrives.<\/p>\n\n\n\n It is ambiguous.<\/p>\n\n\n\n The team interprets the ambiguity generously because everyone likes the idea.<\/p>\n\n\n\n Another month is added.<\/p>\n\n\n\n More money is allocated.<\/p>\n\n\n\n The campaign is given another chance.<\/p>\n\n\n\n That is not validation.<\/p>\n\n\n\n That is delayed decision-making.<\/p>\n\n\n\n The first requirement is a clear statement of what you believe will happen.<\/p>\n\n\n\n Not:<\/p>\n\n\n\n \u201cWe think customers will like this.\u201d<\/em><\/p>\n\n\n\n Something more precise:<\/p>\n\n\n\n \u201cWe believe mid-sized professional-services firms will book a diagnostic session when the offer is positioned around predictable growth rather than marketing execution.\u201d<\/em><\/p>\n\n\n\n Now the assumption can actually be tested.<\/p>\n\n\n\n A useful hypothesis must be capable of being proven wrong.<\/p>\n\n\n\n The second requirement is deciding what the smallest reasonable investment is that can produce a meaningful signal.<\/p>\n\n\n\n A test that is too small may not tell you anything.<\/p>\n\n\n\n You might conclude that the offer does not work when the real problem is simply that nobody saw it.<\/p>\n\n\n\n The objective is not necessarily to achieve statistical certainty in every SME experiment. The objective is to spend enough time and money to generate a useful signal without committing the resources required for a full-scale rollout.<\/p>\n\n\n\n The right threshold depends on the business, channel, audience and decision being made.<\/p>\n\n\n\n This is the part businesses most often avoid.<\/p>\n\n\n\n Before the experiment starts, decide what result would cause you to stop.<\/p>\n\n\n\n Not \u201creview.\u201d<\/p>\n\n\n\n Not \u201cgive it another month.\u201d<\/p>\n\n\n\n Not \u201ctry a different creative.\u201d<\/p>\n\n\n\n Stop.<\/strong><\/p>\n\n\n\n For example:<\/p>\n\n\n\n If the campaign does not generate qualified leads below the agreed acquisition threshold after the defined test period, we stop the campaign and reassess the assumption.<\/em><\/p>\n\n\n\n The exact number will vary.<\/p>\n\n\n\n What matters is that the threshold exists before you see the result.<\/p>\n\n\n\n Otherwise, the result will influence the rule instead of the rule influencing the decision.<\/p>\n\n\n\n There is a human element to this.<\/p>\n\n\n\n Founders become attached to ideas. Teams become attached to work they have spent weeks creating. Leadership becomes attached to decisions they have already approved.<\/p>\n\n\n\n Once money, time and reputation have been invested, stopping can feel like admitting that the original decision was wrong.<\/p>\n\n\n\n But the money already spent is gone.<\/p>\n\n\n\n The relevant decision is not whether the business was right six weeks ago.<\/p>\n\n\n\n The relevant decision is:<\/p>\n\n\n\n \u201cKnowing what we know now, should we spend the next naira on this?\u201d<\/strong><\/p>\n\n\n\n That is the discipline validation creates.<\/p>\n\n\n\n Imagine a company has spent \u20a61.5 million developing and promoting a new offer.<\/p>\n\n\n\n The campaign produces weak results.<\/p>\n\n\n\n The natural reaction may be:<\/p>\n\n\n\n \u201cWe have already spent \u20a61.5 million. Let’s give it another \u20a6500,000 so we can make it work.\u201d<\/em><\/p>\n\n\n\n That logic is dangerous.<\/p>\n\n\n\n The \u20a61.5 million should have no influence on whether the next \u20a6500,000 is a good investment.<\/p>\n\n\n\n The only question that matters is whether the remaining investment has a reasonable probability of producing a worthwhile return based on what the business now knows.<\/p>\n\n\n\n Sometimes the answer will be yes.<\/p>\n\n\n\n Often, it should be no.<\/p>\n\n\n\n The purpose of a kill criterion is to make that decision easier before emotional attachment takes over.<\/p>\n\n\n\n A good validation process should not be a one-off event.<\/p>\n\n\n\n It should become part of how the business operates.<\/p>\n\n\n\n Before launching a major initiative, ask what can be tested first.<\/p>\n\n\n\n These small tests create evidence.<\/p>\n\n\n\n And evidence reduces the cost of being wrong.<\/p>\n\n\n\n A campaign running is not evidence that the idea is working.<\/p>\n\n\n\n A lot of traffic is not necessarily evidence that the offer is working.<\/p>\n\n\n\n High engagement is not necessarily evidence that customers will buy.<\/p>\n\n\n\n Positive feedback is not necessarily evidence of willingness to pay.<\/p>\n\n\n\n People saying they like an idea and people actually changing their behaviour are two different things.<\/p>\n\n\n\n The closer your test gets to the real commercial behaviour you care about, the stronger the evidence becomes.<\/p>\n\n\n\n If the business needs customers to pay, a paid transaction is stronger evidence than a compliment.<\/p>\n\n\n\n If the objective is qualified lead generation, a qualified lead is more useful than a large number of impressions.<\/p>\n\n\n\n If the objective is retention, repeat behaviour matters more than an initial purchase.<\/p>\n\n\n\n Validation should measure behaviour whenever possible.<\/p>\n\n\n\n A practical validation cycle can be simple:<\/p>\n\n\n\n The important part is the final step.<\/p>\n\n\n\n Testing without a decision rule creates information without action.<\/p>\n\n\n\n The purpose of validation is not to create reports.<\/p>\n\n\n\n It is to create better decisions.<\/p>\n\n\n\n Every meaningful experiment should eventually lead to one of three decisions.<\/p>\n\n\n\n The evidence does not support the assumption strongly enough to justify further investment.<\/p>\n\n\n\n Stop.<\/p>\n\n\n\n The underlying idea may still have potential, but an important part of the proposition is not working.<\/p>\n\n\n\n Change the variable you believe is responsible and test again.<\/p>\n\n\n\n The experiment has produced enough evidence to justify increasing the investment.<\/p>\n\n\n\n Scale deliberately, rather than simply switching from a small test to unlimited spending.<\/p>\n\n\n\n Scaling is still another test.<\/p>\n\n\n\n The real value of validation is not just saving money.<\/p>\n\n\n\n It protects management attention. It protects team capacity. It protects operational resources.<\/p>\n\n\n\n Most importantly, it protects the business from spending an entire quarter executing something that should have been challenged in week two.<\/p>\n\n\n\n A bad idea that is killed quickly is not a failure.<\/p>\n\n\n\n It is useful information purchased cheaply.<\/p>\n\n\n\n A bad idea that survives for six months because nobody wanted to stop it is an expensive management failure.<\/p>\n\n\n\n Before committing significantly more money to a new strategy, campaign, product or market, ask:<\/p>\n\n\n\n If the team cannot answer the fourth question, the experiment is not ready.<\/p>\n\n\n\n Strategy is full of assumptions.<\/p>\n\n\n\n You assume a customer will respond.<\/p>\n\n\n\n You assume a message will resonate.<\/p>\n\n\n\n You assume a price will work.<\/p>\n\n\n\n You assume a channel will convert.<\/p>\n\n\n\n You assume a new market will be attractive.<\/p>\n\n\n\n The problem is not having assumptions.<\/p>\n\n\n\n The problem is treating assumptions as facts.<\/p>\n\n\n\n Validation creates the discipline to separate the two.<\/p>\n\n\n\n The objective is not to eliminate uncertainty. That is impossible.<\/p>\n\n\n\n The objective is to reduce uncertainty before the cost of discovering the truth becomes too high.<\/p>\n\n\n\n That is what makes experimentation a core part of a serious Growth System<\/a>.<\/p>\n\n\n\n Business validation is the process of testing an important assumption with real evidence before committing significant resources to it. The goal is to determine whether an idea deserves to be changed, stopped or scaled.<\/p>\n\n\n\n Validation allows SMEs to identify weak assumptions before they consume a full campaign budget, quarter or operational commitment. Small tests can provide useful evidence at a much lower cost than full-scale failure.<\/p>\n\n\n\n A kill criterion is a predefined threshold that determines when an experiment should be stopped. It is agreed before the test begins so the team cannot keep moving the goalposts after seeing an ambiguous result.<\/p>\n\n\n\n There is no universal amount. The test should use the smallest reasonable investment capable of producing a meaningful signal for the specific decision being made.<\/p>\n\n\n\n A business should scale when the experiment produces sufficient evidence that the underlying assumption is commercially promising and the economics support further investment. Scaling should still be measured and controlled rather than treated as an automatic commitment to unlimited spending.<\/p>\n\n\n\n Validation tells you whether an idea deserves more investment. The next challenge is making sure the winning idea can actually be delivered consistently when the business increases its reach.<\/p>\n\n\n\n That means moving from a tested proposition to reliable execution, clear operational handoffs and a launch process that does not introduce new weaknesses at the point of scale.<\/p>\n\n\n\n You can explore more growth strategy and execution insights through the Elan Africa Growth Lab<\/a>.<\/p>\n\n\n\n If you want to assess where your business currently stands before making another major growth investment, take the Growth Diagnostic<\/a>.<\/p>\n\n\n\nValidation Is Not Caution. It Is Cheap Information.<\/h2>\n\n\n\n
What Validation Actually Means<\/h2>\n\n\n\n
\n
The Connection Between Diagnosis and Validation<\/h2>\n\n\n\n
Designing an Experiment That Can Actually Fail<\/h2>\n\n\n\n
The Three Things Every Experiment Needs<\/h2>\n\n\n\n
1. A Specific Hypothesis<\/h3>\n\n\n\n
2. A Minimum Viable Spend<\/h3>\n\n\n\n
3. A Kill Criterion<\/h3>\n\n\n\n
Why Founders Keep Bad Ideas Alive<\/h2>\n\n\n\n
The Sunk Cost Trap<\/h2>\n\n\n\n
Validation Should Be Cheap Enough to Repeat<\/h2>\n\n\n\n
\n
Do Not Confuse Activity With Validation<\/h2>\n\n\n\n
What a Strong Validation Cycle Looks Like<\/h2>\n\n\n\n
\n
Kill, Change or Scale<\/h2>\n\n\n\n
Kill<\/h3>\n\n\n\n
Change<\/h3>\n\n\n\n
Scale<\/h3>\n\n\n\n
Validation Protects the Quarter<\/h2>\n\n\n\n
Five Questions Before You Scale<\/h2>\n\n\n\n
\n
Validation Is How Strategy Becomes Evidence<\/h2>\n\n\n\n
Frequently Asked Questions<\/h2>\n\n\n\n
What is validation in business strategy?<\/h3>\n\n\n\n
Why should SMEs validate ideas before launching them at full scale?<\/h3>\n\n\n\n
What is a kill criterion?<\/h3>\n\n\n\n
How much should a business spend on a validation test?<\/h3>\n\n\n\n
When should a business scale an experiment?<\/h3>\n\n\n\n
What Comes Next?<\/h2>\n\n\n\n