BeginnerLesson 3 of 305 min read

The Marketing Funnel: From Noticing to Buying

Understand the funnel from awareness to loyalty, measure the share that moves on at each step with a worked example, and decide which stage to improve first.

In this lesson you will

  • Name the stages of a funnel
  • Work out the share that moves on at each step
  • Compare a percentage point with a relative gain
  • Choose which stage to improve first

About 22 minutes

Where do the people go?

Noor looks at her month. Twenty thousand people saw a post or an ad. Twelve hundred visited her site. A hundred and eighty joined her email list, twenty-seven bought, and six of those bought again. It is tempting to look only at the 27 and feel disappointed. But each number is a doorway, and people drop out between doorways. A marketing funnel is just that picture: a wide top where many people notice you, narrowing to the few who buy and the fewer who return. Knowing where people drop out tells you what to fix first.

The stages

  • Awareness. They see or hear that you exist.
  • Interest. They visit, read, watch, or follow.
  • Consideration. They join the list, compare, ask a question.
  • Purchase. They buy.
  • Loyalty. They return, recommend, buy again.

Real people do not move in a straight line. They go back and forth and skip steps. The funnel is a thinking tool, not a law of nature, but it is very useful because each stage has its own job and its own number.

Measure each doorway

The program below takes Noor’s made-up month, works out what share moved on at each step, finds the smallest share, and then asks a clever question: what happens if each step improves a little? It tries two kinds of improvement, one extra percentage point and a ten per cent relative improvement, so you can compare them.

# MADE-UP month of numbers for the shop
stages = [("saw a post or ad", 20000), ("visited the site", 1200),
          ("joined the email list", 180), ("bought", 27), ("bought again", 6)]

rates = []
for (a, na), (b, nb) in zip(stages, stages[1:]):
    r = nb / na
    rates.append(r)
    print(f"{b}: {100 * r:.1f}% of the step before")

worst = min(range(len(rates)), key=lambda i: rates[i])
print("smallest share:", stages[worst + 1][0])
print("overall: 1 buyer per", round(stages[0][1] / stages[3][1]), "people reached")

def buyers(rs):
    n = stages[0][1]
    for r in rs[:3]:
        n *= r
    return n

base = buyers(rates)
print("buyers now:", round(base, 1))
for i, name in enumerate(["visit", "join", "buy"]):
    rs = list(rates)
    rs[i] += 0.01
    print(f"+1 point at {name}: {buyers(rs):.1f}")
    rs = list(rates)
    rs[i] *= 1.10
    print(f"+10% at {name}: {buyers(rs):.1f}")
visited the site: 6.0% of the step before
joined the email list: 15.0% of the step before
bought: 15.0% of the step before
bought again: 22.2% of the step before
smallest share: visited the site
overall: 1 buyer per 741 people reached
buyers now: 27.0
+1 point at visit: 31.5
+10% at visit: 29.7
+1 point at join: 28.8
+10% at join: 29.7
+1 point at buy: 28.8
+10% at buy: 29.7

Only 6 per cent of the people reached went on to visit, the smallest share, which is normal at the top of a funnel. Overall there was one buyer for every 741 people reached, and the funnel gave 27.0 buyers. Now look at the experiments. A ten per cent relative gain at any stage gives the same 29.7 buyers, because the stages multiply. But one extra point is different: adding a point to the visit step gives 31.5 buyers, while adding a point to the join step gives only 28.8. The difference is the base each point applies to, since the visit rate starts at 6 per cent and the others at 15, so a point is a bigger proportional gain at the step with the smaller rate. Whichever you improve, the question is the same: which stage can you improve most easily, and which is leaking people you could keep?

What to do with a funnel

  1. Write down your own stage numbers for last month, even if they are estimates.
  2. Work out the share that moved on at each step.
  3. Find the step that looks furthest from normal for your type of business. Ask real customers why people stop there.
  4. Change one thing at that step, such as the headline, the price display or the number of form fields.
  5. Measure again after a few weeks. Do not change five things at once.
Common mistake: always chasing more traffic

Fixing the top first because it is the biggest number. More awareness with a leaky middle just pours more water into the same bucket. Look for the stage where a change is cheap and the loss is real.

Count the same way each time

Percentages at each stage depend on how you count. Be consistent about who counts as a visitor or a buyer, and compare like with like from month to month.

Draw your own funnel with five numbers from last month. Work out the share at each step. Circle the one you would test first and write why.

Practice for this lesson

Check what you learned

Five quick questions. Pick an answer and you will see straight away whether it is right and why.

What is a marketing funnel?

  1. A type of advert
  2. A picture of people narrowing from noticing you to buying
  3. A website tool
  4. A price list

It shows how many people drop out at each stage.

Which stage is first?

  1. Purchase
  2. Loyalty
  3. Consideration
  4. Awareness

People must first know you exist.

Why do small gains at any stage give the same result for a relative gain?

  1. They are added
  2. The stages multiply
  3. Only the top counts
  4. They are unrelated

Buyers equal reach times each stage’s share, so equal relative gains match.

Why can one extra point differ by stage?

  1. Points are random
  2. Only the last stage matters
  3. It never differs
  4. The base each point applies to differs

A point is a larger relative gain on a small rate.

What should you do after finding a leaky step?

  1. Change one thing there and measure again
  2. Change five things
  3. Ignore it
  4. Add more ads at the top

One change at a time shows what helped.

Summary

  • A funnel shows how people narrow from noticing to buying to returning.
  • Measure the share moving on at each step.
  • Stages multiply, so equal relative gains have equal effects.
  • Change one thing at a time at the leakiest step you can fix.

Next lesson

Lesson 4 covers content marketing.

Questions

Is the funnel always accurate?

No. People do not move in a straight line. It is a thinking tool that helps you ask where people stop.

What is a good conversion rate?

It depends on your product, price and traffic type, so compare your own numbers month to month rather than chasing a figure from elsewhere.

Are the numbers real?

No. They are made up to show the arithmetic. Use your own.

Progress is saved in this browser.