AdvancedLesson 22 of 305 min read

Productised Services and Learning Curves

Turn repeated work into fixed packages, price them from honest hours and see how a learning curve raises pay per hour on repeat jobs, with a model on sample numbers.

In this lesson you will

  • Design clear packages
  • Price them from honest hours
  • Model a learning curve
  • Avoid the pitfalls of packages

About 25 minutes

Dev is tired of writing quotes

Dev has now written dozens of quotes, and they are nearly identical: a few pages, a style sheet, a turnaround time. Each one takes time to write, and each client asks the same questions. A productised service is a service sold as a fixed package with a fixed scope, price and delivery time, like a product on a shelf. Instead of ‘tell me what you need and I will quote’, you say ‘here are three options’. It saves time, makes the choice easy for clients and, as this lesson shows, can raise your pay per hour as you get faster.

What makes a good package

  • A clear scope. Exactly what is included, in a short list.
  • A fixed price. No hourly meter, no surprise.
  • A set turnaround. ‘Three working days from approval.’
  • A limit on changes. One or two rounds, as in lesson 6.
  • A stated end. What the client receives, and what is not included.

Three packages are common: a small one for a quick check, a middle one that most people choose and a larger one for bigger needs. The middle option should be the one you most want to sell.

Price packages from your hours

The program below sets up three made-up packages, Check, Polish and Launch, each with a price and an estimate of hours for the first job. Then it applies a learning curve, an idea from manufacturing and many crafts: the more times you do the same job, the faster you get. In this model, every time the number of jobs doubles, the hours fall by 10 per cent. The 10 per cent is made up and your real figure will differ.

import math

# MADE-UP packages: (name, price, hours the first time)
packages = [("Check", 120, 4.0), ("Polish", 295, 10.0), ("Launch", 520, 18.0)]
floor = 24.55
for name, price, hours in packages:
    print(f"{name}: {price / hours:.2f} an hour on the first job")

# a learning curve: each time the number of jobs doubles, hours fall by 10%
rate = math.log2(0.90)
def hours_for(n, first):
    return first * n ** rate
print("Polish hours on the 1st, 2nd, 4th, 8th job:",
      [round(hours_for(n, 10.0), 2) for n in (1, 2, 4, 8)])
for n in (1, 2, 4, 8):
    print(f"job {n}: {295 / hours_for(n, 10.0):.2f} an hour")

per_month = 66 / hours_for(8, 10.0)
print("Polish jobs a month in 66 hours, at job 8 speed:", int(per_month))
print("income from those:", int(per_month) * 295)
Check: 30.00 an hour on the first job
Polish: 29.50 an hour on the first job
Launch: 28.89 an hour on the first job
Polish hours on the 1st, 2nd, 4th, 8th job: [10.0, 9.0, 8.1, 7.29]
job 1: 29.50 an hour
job 2: 32.78 an hour
job 4: 36.42 an hour
job 8: 40.47 an hour
Polish jobs a month in 66 hours, at job 8 speed: 9
income from those: 2655

On the first job, the packages pay 30.00, 29.50 and 28.89 an hour, all above the floor of 24.55. Now follow the Polish package through repeat jobs. The hours drop to [10.0, 9.0, 8.1, 7.29] for the 1st, 2nd, 4th and 8th job. The pay per hour rises from 29.50 at job 1 to 32.78 at job 2, 36.42 at job 4 and 40.47 at job 8. Nothing about the price changed. Dev just got faster. With about 9 jobs a month at job 8 speed, income from this package alone comes to 2655 from 66 billable hours. Compare that with only about 5.5 jobs a month on a custom-quoted 12-hour job. Productising does not guarantee more work. It simply lets the same hours produce more of the same output.

Build your own packages

  1. List the three jobs you do most often and write what each one includes.
  2. Estimate hours from your time log, and add a buffer, as in lessons 7 and 16.
  3. Price each package at or above your floor, and check the pay per hour.
  4. Write a one-page description for each, with scope, price, turnaround and changes.
  5. Offer a way to add extras, quoted separately, so the packages stay simple.

Pitfalls of packages

  • Too rigid. Some clients need something different. Keep room for custom quotes.
  • Hidden extras. If clients often need the same add-on, it belongs in the package.
  • Learning curves stop. After many jobs, the gains are small, so do not plan on them forever.
  • Quality slips. Speed is good only if the work stays good. Use a checklist, as lesson 9 suggested.
  • Underpricing early. You may be slower at first, so price for your honest hours, not your best future ones.

Packages and your reputation

A clear package is easy to refer. A client can tell a friend, ‘she does the Polish package, it is quick and clear’. It also makes your site, proposals and invoices simpler, because the words are already written. Put the packages on a single page, with a short example of the result and a clear next step.

Common mistake: pricing for future speed

Pricing a package for the speed you hope to reach in a year. At first you will be slower and the pay per hour lower. Price for today’s honest hours, and let the learning curve be a bonus.

Sample data and no promises

The packages, prices, hours and the 10 per cent learning rate are made up in sample money. Real learning rates vary and eventually flatten. No income is promised, and this lesson is not financial advice.

Choose a job you often quote. Write it as a package with scope, price, turnaround and change limit, and work out its pay per hour on your honest first-job hours.

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 productised service?

  1. A free sample
  2. A service sold as a fixed package
  3. A product made in a factory
  4. A discount

Scope, price and delivery are fixed.

What does a learning curve say?

  1. Jobs get slower
  2. Prices fall
  3. Repeating a job makes you faster
  4. Clients leave

Hours fall as the number of repeats grows.

What happened to pay per hour as the same package was repeated?

  1. It rose
  2. It fell
  3. It stayed the same
  4. It vanished

The price stayed fixed while the hours fell.

How should you price a new package?

  1. From your hoped-for future speed
  2. From a rival’s guess
  3. From honest first-job hours
  4. From nothing

Do not price for speed you have not reached.

Why keep room for custom quotes?

  1. Some clients need something different
  2. Packages are illegal
  3. It raises tax
  4. It is a rule

Rigid packages cannot fit every client.

Summary

  • A package has a fixed scope, price, turnaround and change limit.
  • A learning curve raises pay per hour on repeat jobs.
  • Price for honest hours today, not hoped-for speed.
  • Keep room for custom work.

Next lesson

Lesson 23 builds authority with honest case studies.

Questions

How many packages should I offer?

Three is common: small, middle and large. Start with one if that is all you can define clearly.

Does the learning curve always hold?

No. Gains flatten as you get very good, and a different client or topic can slow you down again. The 10 per cent in the lesson is made up.

Can I change my packages later?

Yes. Review them every few months against your time log and your clients' questions.

Progress is saved in this browser.