AdvancedLesson 28 of 305 min read

Professional Life: Selling SQL Work Honestly

Work out a floor rate, price a reporting project and a smaller option, show the client's payback including the weak case and promise only what you control.

In this lesson you will

  • Work out a floor rate
  • Estimate by task with a buffer
  • Show an honest client payback
  • Write an offer that promises only what you control

About 28 minutes

Dev’s first paid reporting job

A small chain of shops asks Dev to turn their messy exports into a monthly sales report. He is delighted, then nervous. What should he charge? What should he promise? What if the report is never used? This lesson is about the professional side of SQL work: pricing from your own needs, showing the client an honest picture and promising only what you control. Every figure here is made up, in sample money, and no market rates are quoted.

Start with your floor

Your floor rate is the lowest hourly figure that covers the pay you need and your costs over the hours you can actually bill. Dev plans for a made-up 22000 a year in pay, 2400 in costs, such as software and a laptop’s share, and 800 billable hours, fewer than he works, because finding clients, admin and learning take time. That makes 30.5 an hour, found by dividing 24400 by 800. Never quote below the floor without deciding, on purpose, that the job is worth it for another reason.

Estimate by task, then add a buffer

SQL projects hide surprises, mostly in the data. Estimate each task separately: requirements, cleaning, queries, checks and handover. Then add a buffer for the unknown, and say plainly what it covers. If you do not use it, you can say so. That honesty earns repeat work.

import math
from fractions import Fraction as F

# MADE-UP planning numbers in sample money
pay, costs, billable = 22000, 2400, 800
floor = F(pay + costs, billable)
print("floor rate:", float(floor), "an hour")

def price(tasks, buffer=F(1, 5)):
    hours = sum(tasks.values())
    planned = hours * (1 + buffer)
    return hours, planned, planned * floor

full = {"requirements": 4, "cleaning": 10, "queries": 12, "checks": 6, "handover": 4}
small = {"cleaning": 8, "queries": 6, "checks": 3, "handover": 2}
net = 6 * 14 - 20          # client saves 6 hours a month at 14, and pays 20 a month for care
for name, tasks in (("full project", full), ("small project", small)):
    hours, planned, fee = price(tasks)
    print(f"{name}: {hours} h, {float(planned)} h with buffer, fee {float(fee):.2f}")
    print(f"  months to pay back: {math.ceil(fee / net)}")
    half = 6 * 14 / 2 - 20
    print(f"  if the saving is half: {math.ceil(fee / half)} months")
print("client net gain a month:", net)
floor rate: 30.5 an hour
full project: 36 h, 43.2 h with buffer, fee 1317.60
  months to pay back: 21
  if the saving is half: 60 months
small project: 19 h, 22.8 h with buffer, fee 695.40
  months to pay back: 11
  if the saving is half: 32 months
client net gain a month: 64

The full project adds up to 36 hours, or 43.2 with a 20 per cent buffer, so the fee at the floor is 1317.60. A smaller version, which leaves out the requirements work and trims the queries, comes to 19 hours, and its fee is 695.40. Now the part most beginners skip: the client’s side. If her estimate is right, the new report saves her 6 hours a month at a value of 14 an hour, which is 84, and she pays 20 a month to keep it working, so her net gain is 64 a month. The full project takes 21 months to pay back. The small one takes 11. If the saving is only half of what she hopes, the full project takes 60 months and the small one 32.

That is an uncomfortable result, and a professional says it out loud. For this client, the full project is hard to justify on savings alone. Dev can offer the small version, or ask whether other benefits matter, such as fewer mistakes or better decisions, and let the client decide. He should never inflate the savings to win the work. Honest numbers lose some jobs and win the clients who come back.

What to put in an offer

  • Scope. What the report contains and what it does not.
  • Data. What you need from the client, such as access and a named person, and what you do if the data is poor.
  • Checks. How you will reconcile totals, as in lessons 25 and 26.
  • Handover. Documentation, the queries and a short walk-through.
  • Care. What happens after delivery, how problems are reported and how quickly you reply.
  • Privacy. Who sees personal data, where it is kept and when it is deleted. This is an outline, not legal advice.

What not to promise

  • A saving or a payback. They are the client’s estimates.
  • That the numbers will be perfect. Promise your checks and your reconciliation instead.
  • That the client’s source data is clean. You can promise to report on its problems.
  • Features of a database product that you have not tested in the client’s own system.

Put the agreement in writing: scope, price, payment dates, the changes process, who owns the work and what happens if either side stops. For anything with real consequences, take proper legal and tax advice for your country.

Common mistake: pricing below your floor by accident

Quoting a low price to win a first job and finding that you earn less than a wage. If the job is worth doing for another reason, such as a portfolio piece, decide that deliberately and say so in your notes. Never discover it afterwards.

Made-up sample data, outline only

The pay, costs, hours, savings and fees are made up, in sample money. No market rates are quoted, and no saving is promised. Contract, tax and privacy points are outlines, not legal, tax or financial advice.

Work out your own floor rate from a pay target, costs and honest billable hours. Price a small imaginary report job, and then show the client payback, including the half-savings case.

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 floor rate?

  1. The average market price
  2. A tax rate
  3. The client’s budget
  4. The lowest hourly figure covering your pay and costs over billable hours

Dividing 24400 by 800 gave 30.5.

Why show the client the half-savings payback?

  1. It lowers your fee
  2. It gives an honest worst case before she decides
  3. It avoids a contract
  4. It hides risk

The full project took 60 months in that case.

Why offer a smaller project?

  1. It pays you more per hour
  2. It removes checks
  3. It avoids a floor rate
  4. It may suit the client’s likely savings better

Its payback was 11 months, not 21.

Which should you not promise?

  1. The client’s savings or payback
  2. Your reconciliation checks
  3. A written scope
  4. A handover walk-through

Savings are the client’s estimate.

What does the buffer cover?

  1. Your profit only
  2. Tax
  3. The unknown, mostly surprises in the data
  4. The care plan

Say plainly what it covers and whether you used it.

Summary

  • Quote at or above your floor rate.
  • Estimate by task and add a stated buffer.
  • Show the payback honestly, including the weak case.
  • Promise your checks, not the client’s savings.

Next lesson

Lesson 29 shows how to handle SQL projects: scope, critical path and change.

Questions

Should I charge by the hour or by the project?

Either can work. A project fee needs a clear scope and a way to price changes. Start from your floor in both cases.

What if the client's data is very poor?

Say so early, price a data review first and report on the problems. Do not promise results from data you have not seen.

Are these figures real?

No. All pay, costs, hours and savings are made up, in sample money.

Progress is saved in this browser.