Up to this point, we have established the basic financial and measurement rules that should exist before a business spends money on Meta advertising. We know what a sale is worth, how much we can afford to spend acquiring that sale, how much money we are willing to risk during the initial test, and what conditions will cause us to continue, change direction, or stop.
Now we can begin the actual strategy.
The method used throughout the remainder of this guide is based on a simple idea: when you are trying to find customers, you do not necessarily know where the customers are before you begin. You may have strong assumptions. You may know the general type of person who should want the product. You may even have previous customer data. But until money is placed into the market and people begin responding, much of what you believe is still a hypothesis.
Wares Point approaches this problem using what we will call the Fishing Pond Method.
The analogy is intentionally simple because the strategy itself should remain simple.
Imagine that you are trying to catch a particular type of fish. You have several ponds available to you, and each pond contains a different population. Some may be full of the fish you want. Some may contain a few. Some may look promising but contain almost none at all.
You do not know with certainty which pond will produce the best result.
That is the same problem an advertiser faces when choosing audiences.
The goal of the Fishing Pond Method is not to guess perfectly before the campaign launches. The goal is to create a controlled environment in which several reasonable possibilities can compete, then allow the results to tell us where the strongest opportunities actually exist.
The Pond Represents the Audience
In this method, each pond represents an audience or market segment that has a logical relationship with the product being advertised.
Suppose, for example, that a business sells a product designed for people starting an online business. There are several reasonable ponds that might contain potential customers. One pond might contain people interested in entrepreneurship. Another might contain people interested in e-commerce. Another could contain affiliate marketers. Another might contain people interested in side hustles, digital products, freelancing, or small-business software.
Every one of these groups appears relevant.
That does not mean they will perform equally.
One of the biggest mistakes in advertising is assuming that because an audience makes sense logically, it must also make sense economically. An audience can appear perfect on paper and still produce expensive, low-quality traffic. Another audience may look less obvious but consistently generate buyers.
The market makes the final decision.
The advertiser's job is to choose several ponds that are reasonable enough to deserve a test.
For the type of initial campaign discussed throughout this guide, Wares Point generally begins with approximately five to seven distinct audience ideas. That is enough variation to compare different possibilities without dividing a modest testing budget across an unreasonable number of directions.
The purpose is not to create every audience that could possibly work.
The purpose is to create a manageable group of good hypotheses.
The Bait Represents the Advertisement and Offer
Finding the right pond is only part of the problem.
A pond can contain plenty of fish and still produce poor results if the bait is wrong.
In advertising, the bait represents the combination of the creative, message, product presentation, and offer being placed in front of the audience.
Two advertisers can target the exact same audience and receive completely different results because their advertisements communicate differently.
One may use an image that immediately communicates the value of the product. Another may use an attractive image that has almost nothing to do with the customer's actual problem.
One may lead with a compelling benefit. Another may open with generic language that gives the viewer no reason to stop scrolling.
One may present a clear $10 offer that feels like an easy decision. Another may make the exact same $10 product seem confusing or untrustworthy.
The audience matters, but so does what the audience sees.
This is why the Fishing Pond Method does not rely on a single advertisement.
A pond should be given more than one opportunity to demonstrate whether it contains customers.
For an initial test, Wares Point may prepare approximately five to ten relevant creative variations that can be used across the campaign. These do not need to be five to ten completely different products or offers. They may represent different images, headlines, opening hooks, angles, demonstrations, layouts, or presentations of the same underlying offer.
The objective is to avoid reaching a false conclusion such as:
This audience does not work.
when the real conclusion should have been:
This particular advertisement did not work with this audience.
That distinction becomes increasingly important as campaign data begins to accumulate.
The Advertising Budget Is the Fish Food
The third part of the analogy is the money itself.
Advertising spend is the fish food being placed into the pond.
If you put nothing into the water, you learn nothing.
If you dump your entire supply into one pond immediately, you may discover that the pond was a poor choice only after most of the budget has disappeared.
The Fishing Pond Method therefore begins with controlled feeding.
For the example used throughout this book, we have been working with an initial advertising budget of approximately $25 per day.
That $25 is not expected to prove the entire business model overnight. It is the budget assigned to the first stage of discovery.
The objective is to give the campaign enough money to begin revealing patterns without exposing the business to unlimited loss.
This is also where campaign-level budgeting becomes useful.
When a shared campaign budget is used, Meta is given some freedom to distribute that daily budget among the ad sets based on the opportunities its delivery system detects. Advertisers commonly refer to this structure as CBO, or Campaign Budget Optimization, although Meta's interface may use newer Advantage+ campaign budget terminology.
Within the Fishing Pond Method, the idea is straightforward: several ponds are placed inside the same controlled experiment, and the system begins determining where the available money can be spent most effectively.
That does not mean we blindly accept every decision Meta makes.
It means we allow the campaign to generate evidence before we begin interfering with it.
Why We Start With Several Ponds
A beginner often wants to find the one perfect audience before spending anything.
That is understandable, but it places too much confidence in assumptions.
Imagine choosing a single audience because it seems ideal, spending the entire test budget against it, receiving mediocre results, and concluding that Meta advertising does not work for the product.
The advertiser has actually learned very little.
Perhaps the product could have performed extremely well with another audience that was never tested.
The Fishing Pond Method deliberately avoids placing the entire experiment into one pond.
Instead, we begin with several audiences that have clear strategic reasons for being included.
For example, a five-pond test might contain five different customer hypotheses. The first pond could represent the most obvious buyer. The second might represent a closely related interest group. The third could represent people attracted to the desired outcome rather than the product category itself. The fourth may represent a complementary industry. The fifth may test a broader audience than the advertiser would normally consider.
Each pond should have a reason to exist.
Random audiences do not create useful experiments.
We are not asking, "What audiences can we possibly target?"
We are asking:
What are the most plausible groups of people who could reasonably want what we are selling?
The First Three Days Are an Observation Period
Once the campaign launches, one of the hardest things for a new advertiser to do is leave it alone.
The advertiser begins checking results almost immediately.
One audience receives more spending than another, so the advertiser changes the budget.
An advertisement receives several clicks but no sale within a few hours, so it is turned off.
Another ad receives no impressions yet, so it is edited.
The campaign is changed repeatedly before it has had enough time to produce meaningful information.
Within the Fishing Pond Method, the initial launch is treated as an observation period.
For a small campaign, Wares Point generally uses approximately three days as the first major observation window.
This is not a magical number and should not be treated as a universal Meta rule. It is simply a practical discipline for preventing premature interference while a modest campaign begins accumulating information.
During this period, the advertiser is still watching the campaign.
We are checking for obvious technical failures. We want to know that advertisements are actually delivering, links work, the website loads, purchases can be tracked, and nothing is consuming money in a clearly abnormal way.
What we are trying to avoid is constant strategic intervention.
If the campaign is structurally sound, let the ponds receive some food.
Let the advertisements enter the market.
Let people respond.
Then examine what happened.
Do Not Expect Every Pond to Receive the Same Amount of Money
When using a shared campaign budget, the advertiser should not assume that every audience will receive exactly the same amount of spend.
That would defeat much of the purpose of allowing campaign-level optimization.
Meta may begin directing more money toward one ad set and less toward another based on the signals it is observing.
That behavior itself becomes useful information.
Suppose a campaign contains six ponds.
After several days, Meta has spent much more heavily in two of them, moderately in another two, and very little in the final two.
That does not automatically prove that the first two ponds contain the most profitable customers. Delivery decisions can be influenced by many factors, and actual purchases remain more important than spend alone.
However, it gives us something to investigate.
We begin looking for combinations of three things:
Where is the money being spent?
Where are people meaningfully responding?
Where are purchases actually occurring?
Those three questions are much more useful than simply asking which audience has the cheapest clicks.
A Click Is Not a Fish
This distinction is central to the Fishing Pond Method.
Clicks are activity.
Purchases are the result we are ultimately trying to find.
Imagine one pond produces extremely cheap clicks. People seem interested, traffic flows to the site, and the campaign dashboard looks active.
But nobody purchases.
Another pond produces fewer clicks at a higher cost, but a meaningful percentage of those visitors purchase the product.
Which pond contains the better fish?
For a sales campaign, the second one.
This is why Wares Point does not judge ponds solely by CTR, CPC, or engagement. Those measurements can help explain what is happening, but they do not replace the financial result.
An audience full of people who enjoy clicking advertisements is not automatically a valuable audience.
We are looking for buyers.
Give the Pond More Than One Bait Before You Declare It Empty
Another important principle is avoiding premature judgment.
Suppose Audience A receives Advertisement 1 and performs poorly.
That does not necessarily prove Audience A is bad.
It may mean Advertisement 1 was poor bait.
If another creative communicates the offer differently and suddenly begins producing purchases within the same audience, the pond was never the problem.
The Fishing Pond Method therefore encourages the advertiser to evaluate the relationship between audience and creative, not merely each component in isolation.
Over time, the campaign may reveal that one creative works across several audiences. It may reveal that a particular audience responds strongly to one message and ignores another. It may reveal that several audiences behave similarly enough that they no longer need to be treated separately.
This is where advertising begins moving away from assumption and toward evidence.
Know When to Stop Feeding an Empty Pond
The financial exit plan developed earlier now becomes part of the Fishing Pond Method.
We already determined the maximum amount we can afford to spend acquiring a customer.
That number gives us a reference point for deciding how much money a pond should be allowed to consume without producing a sale.
Suppose the maximum acceptable CPA is $4.
If a particular pond or advertisement spends $1 without a purchase, we have learned very little.
At $4 without a sale, we are paying attention.
At $8, the test has consumed the equivalent budget for two acceptable customer acquisitions without producing one.
At $12, we have spent approximately three times the maximum allowable CPA without obtaining the desired result.
At some point, continuing to feed the exact same empty pond stops being experimentation and begins becoming denial.
This is why Wares Point uses the two-to-three-times maximum CPA range as a serious review threshold for low-priced products.
It is not an automatic kill switch.
If the pond is generating strong checkout activity, for example, the advertiser may investigate whether something later in the sales process is preventing purchases.
If there is virtually no meaningful activity at all, there is much less justification for continuing.
The important point is that the stopping decision was defined before the campaign launched.
We are not deciding how much loss feels acceptable after the money has already disappeared.
Identify the Productive Ponds
After the initial observation period, the campaign should begin telling a story.
Suppose we launched six audiences with a $25 daily campaign budget.
After three days, approximately $75 has entered the system.
The results might look something like this:
Pond
Spend
Purchases
CPA
Audience A
$18
6
$3.00
Audience B
$14
4
$3.50
Audience C
$12
1
$12.00
Audience D
$11
0
—
Audience E
$9
2
$4.50
Audience F
$11
5
$2.20
Now we have something worth analyzing.
Audience F appears particularly interesting.
Audience A is also performing within our target economics.
Audience B may still be useful.
Audience E is producing purchases but sits above the desired acquisition cost.
Audience C appears expensive.
Audience D has not produced a sale.
This does not mean we immediately turn everything off except Audience F.
The sample is still relatively small.
However, the ponds are no longer theoretical.
We now have evidence.
The next stage of campaign management is deciding which ponds deserve more food, which need different bait, and which should stop receiving money.
Winners Earn More Budget; They Are Not Entitled to It
Finding an early winner does not mean the campaign is finished.
Advertising performance changes as spending increases.
An audience producing customers for $2.50 at a modest budget does not automatically continue producing customers for $2.50 after the budget is multiplied several times.
There may be only a certain amount of efficient demand available within that particular combination of audience, creative, and offer.
Scaling therefore happens gradually.
The Fishing Pond Method does not recommend discovering one productive pond and immediately dumping the entire advertising budget into it.
Instead, the business gives the productive pond additional resources and watches what happens.
If sales increase while acquisition cost remains acceptable, the pond earns the right to receive more.
If CPA deteriorates significantly, the business slows down and reassesses.
Scaling is therefore another experiment.
The difference is that the experiment now begins with evidence rather than a guess.
The Method in Practice
The overall Fishing Pond Method can now be summarized as a repeatable process.
Begin by calculating the economics of the product and determining what a customer can affordably cost.
Select approximately five to seven reasonable audience hypotheses.
Prepare several strong creative approaches so that individual audiences are not judged using only one piece of bait.
Place those audiences into a controlled campaign with a predefined daily and total testing budget.
Allow the campaign an initial observation period—approximately three days for the type of small test used in this guide—while watching for technical problems rather than constantly changing strategy.
After the observation period, evaluate which audiences and creatives are producing meaningful customer behavior and actual purchases.
Use the maximum allowable CPA established before launch to identify ponds that are consuming more money than their results justify.
Stop feeding combinations that repeatedly fail.
Continue testing promising combinations.
Give additional resources to audiences and creatives that produce customers within acceptable economics.
Then repeat the process.
This is not a one-time trick.
It is a system for gradually moving advertising money away from assumptions and toward evidence.
What We Are Actually Trying to Discover
The Fishing Pond Method is not designed to prove that the advertiser was right.
It is designed to discover where the advertiser is wrong as cheaply as possible and where the advertiser is right as profitably as possible.
That difference matters.
If your favorite audience performs poorly, the answer is not to keep feeding it until the campaign eventually produces a sale.
If an unexpected audience performs exceptionally well, the answer is not to ignore it because it does not fit the original theory.
The market is providing information.
Use it.
A disciplined advertiser is not emotionally attached to the pond.
The advertiser is attached to the economics.
The Stopping Point for This Stage
At the end of the first Fishing Pond test, every audience should fall into one of three practical categories.
Some ponds will have produced enough evidence to justify continued investment.
Some will remain uncertain and deserve a better piece of bait or a limited additional test.
Others will have consumed enough money without producing sufficient evidence of customer demand and should be stopped.
That is the completion point for the first discovery cycle.
We do not continue testing indefinitely simply because additional audience ideas exist.
The objective of this stage was to identify the most promising places to concentrate the next round of advertising.
Once that has been accomplished, we move forward.
Next: Choosing the Right Ponds — Building the First 5–7 Audiences
The next chapter will focus entirely on audience selection.
We will take a product, identify the types of people most likely to buy it, turn those assumptions into five to seven distinct audience hypotheses, and explain how to avoid creating audiences that are so similar that the campaign is effectively testing the same pond repeatedly.
From that point forward, the book will build the actual Fishing Pond campaign one component at a time.