Back to the on-screen lesson ·
A channel fits when it brings the segment and a measurable share take the next action; trace each channel with its own code and compare cost per booking before spending more.
Paper packet. Every task here also exists on screen, where it is checked automatically; answers written on paper are not assessed by Nydus. When you are back at a device, enter your answers there.
You will tell evidence of channel fit from attention, trace each channel's next actions with its own code, work out booking rates, and decide where the next dollar goes by cost per booking.
You can choose a channel by where the segment decides and compute what a new customer cost through it. Before spending more on a channel, an owner needs evidence that it actually fits, and that evidence has to be collected on purpose.
| Term | What it means |
|---|---|
| Fit | A channel brings the segment and a measurable share take the next action. |
| Next action | The step the offer needs: a booking, an order, a visit. |
| Tracing | Giving a channel something only it carries: a code, a number, a link. |
| Booking rate | Next actions over contacts, times a hundred. |
| Cost per booking | Spend on a channel over the traced bookings it brought. |
| Vanity figure | A count of attention: views, likes, followers. |
Every channel produces numbers, and most of them feel encouraging. The directory listing had 3,000 views. The advert got 120 likes. The newsletter has 600 readers. None of these would look any different if nobody booked at all, so none of them shows fit.
Fit is shown by the next action the offer needs, counted for that channel alone. Bright Home Cleaning's property-manager cards carry their own booking code: 40 people scanned it last month and 9 booked, a booking rate of 9 ÷ 40 × 100 = 22.5 percent. The directory has a different code, and brought 6 bookings. Now the channels can be compared on what matters.
Tracing costs almost nothing: a different code on each card, a second phone number, a question on the booking form ('which code did you use?'). Without it every booking looks the same, and the owner is left guessing which channel to feed.
Another way: table
Attention against action, one month.
| Channel | Attention | Traced bookings |
|---|---|---|
| Directory listing | 3,000 views | 6 |
| Property-manager cards | 40 scans | 9 |
| Social post | 120 likes | 1 |
Another way: steps
With traced bookings, a channel's cost per booking is its spend divided by its bookings, and that is the comparison that should decide where the next month's money goes. A newsletter costing 180 dollars that brought 6 traced bookings cost 30 dollars a booking; a lunch-fair stand costing 200 dollars that brought 10 cost 20. The fair gets the next month, however many more people read the newsletter.
Two cautions. A month is a short record, and a channel with 3 bookings is one good or bad week away from a different answer; before doubling spend, look for the result to hold for a second period. And a channel that brings bookings from outside the segment, cheap ones that do not fit the offer, is not fitting either, however low its cost per booking: fit means the right people taking the next step. Keep the records that let you check both.
Name the next action. For a cleaner it is a booking; for a kitchen, an order; for a pottery shop, a purchase at checkout. Choose one, so every channel is judged on the same step.
Trace each channel. A code printed on the card, a phone number used only on the sign, a web link used only in the newsletter. Record it with every booking, in the same column of the same log.
Count contacts and actions. Scans, calls or visits for each channel, and the bookings that followed.
Work out the booking rate. Bookings over contacts, times a hundred.
Work out the cost per booking. Spend on the channel over its bookings.
Decide. The next dollar goes to the channel with the lower cost per booking, provided its bookings are in the segment and the result has held for more than one period.
Check the answer. A booking rate can never exceed a hundred percent. Cost per booking times bookings must give back the spend. And the bookings from all channels, plus those with no code, must add up to the total in the booking log; if they do not, some bookings were not recorded.
Counting only traced bookings is allowed because the code proves where the booking came from. A booking with no code could have come from any channel, or from none, so it cannot be credited to one.
Dividing by bookings rather than by contacts is needed because the spend was made to win bookings. A call that did not book cost the same to attract, but it brought nothing in, so it belongs in the booking rate, not in the cost per booking.
Waiting for a second period is allowed, and wise, because small counts jump about by chance. Three bookings one month and seven the next can come from the same channel doing the same thing; a decision built on one month can be a decision built on luck.
A tracing code only works if customers use it and staff record it. Make the code short and worth using: 'CARD10 for ten dollars off your first clean' is used far more often than a long reference number with no reason to type it. Put the question on the booking form or the phone script, so nobody has to remember to ask.
Expect some bookings to arrive without a code. A customer who saw the card may search for the business by name and book on the website. Keep those in an 'unknown' line rather than guessing, and if the unknown line grows large, ask at booking as well: 'How did you hear about us?' still helps, even if the answer is rougher than a code.
A channel can bring bookings and still not fit. A discount site might fill a cleaner's calendar with one-time bargain hunters who never book again; a partner might send tenants whose homes need work the business does not do. Look one step further when the records allow: how many of each channel's customers booked a second time, and how many asked for something the business had to decline. A channel whose customers return is worth more than its cost per booking suggests, and one whose customers never return is worth less.
The two figures in this lesson are easy to confuse, and they can point in opposite directions. The booking rate says how well a channel's contacts turn into bookings: of the people who called the sign's number, what share booked. The cost per booking says what each booking cost the business. A channel can have a high booking rate and a high cost, because it brings few contacts at a high price; or a low booking rate and a low cost, because it brings many contacts very cheaply.
Use each for its own job. The booking rate tells the owner whether the message and the offer are working once a person makes contact: a low rate from a channel full of the right people suggests the offer, the price or the way calls are answered needs work. The cost per booking tells the owner where to put money. When the two disagree, money follows the cost per booking, and attention goes to whatever is dragging the rate down.
Both figures need the same record underneath: contacts, bookings and spend for each channel over the same period. Keep all three, and either figure can be worked out again later from the log.
Some channels resist codes. Word of mouth, a sign on the shop itself, or a mention in a local news story bring customers who never see a code. Do not pretend those channels do not exist because they cannot be counted precisely. Instead, use a before-and-after comparison: count bookings for a few weeks before the change and a few weeks after it, with everything else as steady as possible, and treat the difference as a rough estimate of what the channel added.
The estimate will be less certain than a code, because other things change too: the weather, the season, a competitor's sale. Say so beside the figure, and give it less weight when deciding where money goes. Where a cheap way to trace the channel can be found, such as a 'mention the sign for a free safety check' line, use it, and the rough estimate can be replaced by a count.
A food truck in Philadelphia sold lunches at three spots: a university campus, a hospital parking lot and a business district. Its owner paid for posts aimed at each neighborhood and printed flyers for each, and every spot seemed busy. When rising costs forced a choice of two spots, she had no idea which posts and flyers were working.
For one month she gave each spot's flyers and posts their own QR code for a free drink with a first order, and the window staff scanned the code at the order. The campus material brought 140 scans and 35 first orders. The hospital material brought 60 scans and 30 first orders. The business district brought 90 scans and 9.
The spending had been the same, 270 dollars a spot. That came to about 8 dollars a first order at the campus, 9 at the hospital, and 30 in the business district. She dropped the business district's posts, kept a smaller presence there for regulars, and moved the money to the hospital, where night-shift staff also turned out to return several times a week.
Larger businesses buy call-tracking services that assign a different phone number to each advert, and web tools that tag each link with the campaign it came from. The idea is the same as a code on a card: something only one channel carries, recorded at the moment of the next action. A small business can do it with a second phone line and a column in a log.
Reach proves fit. Views and readers count attention; fit is counted in next actions.
A channel fits because a competitor uses it. They may be reaching a different segment, or wasting their money.
Customers will remember how they heard. Memory is poor; a code at booking is reliable.
Spend more where more was spent. Spend more where a booking cost less.
One good month settles it. Small counts move by chance; look for a second.
Note the attention.
$3000 \text{ views}$
Encouraging, but not fit.
Name the next action.
$\text{a booked clean}$
What the offer needs.
Count traced bookings.
$6 \text{ with the directory's code}$
Credited to this channel alone.
Work out the cost per booking.
$240 \div 6 = 40$
Spend over traced bookings.
Compare with the cards.
$90 \div 9 = 10$
A quarter of the cost per booking.
Note the temptation.
$\text{lots of comments; double the posts?}$
Comments are attention.
Trace the posts.
$\text{their own phone number for a month}$
Makes the next action countable.
Work out the posts' booking rate.
$15 \div 50 \times 100 = 30$
Fifteen repairs from fifty calls.
Work out the sign's booking rate.
$12 \div 20 \times 100 = 60$
Twelve from twenty.
Work out each cost per booking.
$300 \div 15 = 20; \ 60 \div 12 = 5$
Spend over bookings.
Decide where the money goes.
$\text{not more posts}$
The sign wins bookings at a quarter of the cost.
Name the next action.
$\text{a catering order}$
One step for both channels.
Trace each channel.
$\text{code NEWS10; code FAIR10}$
One code each.
Work out the newsletter's cost.
$180 \div 6 = 30$
Spend over traced orders.
Work out the fair's cost.
$200 \div 10 = 20$
Spend over traced orders.
Check the counts are not tiny.
$6 \text{ and } 10$
Small; wait for a second month.
Look again after a second month.
$22 \text{ and } 18 \text{ dollars}$
The fair still cheaper.
Decide where the next month's money goes.
$\text{the fair}$
However many people read the newsletter.
Name the next action the workshop should lead to.
$\text{an order from the online shop}$
The step her offer needs.
Trace the workshop.
$\text{a code on the handout, recorded at checkout}$
Only the workshop carries it.
Work out the cost per order: 120 dollars, 8 coded orders.
Bright Home Cleaning is thinking of doubling what it pays to be listed in a local services directory. Which piece of evidence shows the directory fits the offer?
Complete the worked solution: Northside Repairs gave its cycle-path sign its own phone number. This month the sign cost $72$ dollars and brought $20$ calls, of which $6$ booked a repair. Find the calls that did not book, the booking rate as a percentage, and what each booking cost.
Find the calls that did not book.
$(\text{calls}) - (\text{booked}) =$ x
Attention that did not become a next action.
Find the booking rate.
$(\text{booked}) \div (\text{calls}) \times \text{a hundred} =$ r
The share who took the next step.
Find what each booking cost.
$(\text{spend}) \div (\text{booked}) =$ c
The figure that decides where the next dollar goes.
Say what made these figures possible.
$\text{a number only the sign carries}$
Without it every call looks the same.
Neighborhood Kitchen's report on its office park newsletter advert. Mark every figure that shows the channel fits.
This task has no paper form; do it on a device.
Neighborhood Kitchen spent $189$ dollars on the office newsletter, which brought $9$ traced catering bookings, and $216$ dollars on a lunch-fair stand, which brought $8$. It has money for one more month of one channel. Which should it be?
The card Bright Home Cleaning leaves with property managers has its own booking code. $40$ people scanned it last month and $6$ of them booked a move-out clean. What percentage of those who scanned it booked?
Of those who scanned the card's code, v percent booked.
Northside Repairs gives each channel its own phone number. Last month the forum posts brought $50$ calls and $13$ booked repairs; the cycle-path sign brought $20$ calls and $8$ booked. Fill in each channel's booking rate, as a percentage.
| Booking rate, percent | |
|---|---|
| Forum posts | |
| Cycle-path sign |
A plumber in Houston put yard signs with their own phone number outside finished jobs. Printing the signs cost $56$ dollars and placing them cost $20$ dollars. The number took $12$ calls, and $4$ of them booked a job. What did each booking cost, in dollars?
Answer:
A yoga teacher advertises her classes through $3$ channels: flyers in cafés, a gym noticeboard and a community website. How can she find out which channel brings people who actually book?
Lesson test: one question per skill, one attempt each, no hints. Your answers are checked when you submit.
Bright Home Cleaning spent $196$ dollars on a services directory, which brought $7$ traced bookings, and $148$ dollars on cards left with property managers, which brought $4$. Fill in what a booking cost through each channel, in dollars.
| Cost per booking, dollars | |
|---|---|
| Services directory | |
| Property-manager cards |
You can show whether a channel fits from traced next actions, and choose between two channels on cost per booking. Tell someone why 3,000 views can mean less than 40 scans. Next: the message a channel carries.
18. Your turn: Maya's workshop evening, step 3
$120 \div 8 = 15$
Spend over traced orders.