Module 08 of 11
Keep the Customers You Win
The cheapest growth available, why client relationships fail on communication rather than quality, and what to do with the portfolio nobody is paying you for.
Learning objectives
By the end of this module you will be able to:
- Explain why retention is cheaper than acquisition and what that changes about where you spend effort.
- Match a loyalty tactic to the type of customer it suits.
- Name the communication habits that keep client relationships intact.
- Describe how satisfaction is measured, and the limits of each measure.
- Decide what a portfolio is for when you are not selling anything.
Lesson
Everything up to here has been about being found and being chosen. This module is about the part that decides whether any of it was worth doing: whether the people who chose you once do it again.
The arithmetic of repeat business
Winning a new customer costs money and time — the ad, the proposal, the calls, the negotiation. Serving an existing one who already trusts you skips most of that. The second project with the same client is more profitable than the first even at the same price, because the cost of getting it was a conversation rather than a campaign.
That has a practical consequence most small businesses ignore: an hour spent on a past client is usually worth more than an hour spent on a stranger, and almost nobody allocates their time that way. Marketing is more visibly productive, which is exactly why it wins the hour it should not.
Loyalty tactics are not one thing
The tactic has to fit the customer, and the commonest mistake is applying a high-frequency retail tactic to a low-frequency professional relationship. Points do not work on someone who hires you twice a year.
- Frequent, low-value purchases — a reason to come back that accumulates: a card, points, a subscription, a members’ price.
- Infrequent, high-value purchases — proactive contact, remembering context, and being the person they think of first. No scheme; a relationship.
- One-off purchases with a long life — support that continues after the sale, and useful contact when something changes that affects them.
- Subscribers — onboarding that gets them to the point of value quickly, then not going quiet between renewals.
- People who bought once and vanished — one honest, specific message asking whether they still need what you do, and taking silence as an answer.
Client relationships fail on communication first
Long-running relationships rarely end over the work being bad. They end over silence, surprises and the sense that something was hidden. Four habits cover most of it.
- Say when things will happen, and say something when they will not. A delay announced in advance is a fact; a delay discovered by the client is a character flaw.
- Put money in writing before the work, not after. Almost every serious dispute is a scope conversation that never happened.
- Report progress on a rhythm the client did not have to ask for. Silence is read as trouble, and reading it that way is reasonable.
- When something goes wrong, be the one who says so first. The cost of a mistake you reported is a fraction of the cost of a mistake they found.
Measuring satisfaction, and the limits of it
You can measure satisfaction, but every measure is a proxy and each fails differently. Knowing how they fail is more useful than the numbers.
- Net Promoter Score — "how likely are you to recommend us". Widely used, easy to compare, and easy to game; a single number that hides why.
- Customer satisfaction score — asked right after an interaction, so it captures the moment rather than the relationship.
- Repeat purchase rate — behaviour rather than opinion, and therefore the honest one, but slow.
- Churn — the share who leave in a period. The clearest signal, arriving after it is too late to act on that customer.
- What they actually say — the least structured, hardest to chart, and usually the only one that tells you what to change.
A portfolio when nothing is for sale
Portfolio work has a different job when you are not monetising: it is a record of what you can do, an argument that you finish things, and — most usefully — a set of concrete artefacts to point at in conversations that have not happened yet.
What makes a portfolio piece work is not the polish of the screenshot. It is that it states the problem, what you decided, and what happened as a result. A project with no stated problem is decoration; a project with a problem, a decision and an outcome is evidence.
Assignment
Turn one past client into a second project
Pick a real client or contact you worked with more than six months ago and have not spoken to since. You are going to plan the re-approach properly rather than sending a "just checking in" message.
- Write down what you actually know about their situation, and what has probably changed since you finished.
- Identify one specific, useful thing you could tell them that is true whether or not they hire you — a risk, an expiry, something that has moved on since you built it.
- Draft the message around that thing. It should be sendable even if they never reply, and it should not contain the phrase "just checking in".
- Decide what you will do if there is no reply, and when. Then write that date down.
Deliverable
The drafted message, the specific useful thing it leads with, and the date you have set for the follow-up.
Quiz
Match each type of customer to the loyalty tactic that suits them.
Pick the matching item for each row.
- Buys something small most weeks
- Hires you for a large project roughly once a year
- Bought one thing that will last a decade
- Pays monthly and has not logged in for five weeks
A client had one site built by you eighteen months ago and has not been in touch. What is the strongest way to reopen the conversation?
You will miss a deadline by a week. When do you tell the client?
Which of these measure behaviour rather than stated opinion?
Select every answer that applies.
True or false: a portfolio piece works best when it leads with how the finished thing looks.
Of the last ten hours you spent on getting work, how many went to people who have already paid you once? If the answer is zero, what would you move?
Score: 0 / 5
Recap
Before you move on, the things worth keeping:
- Retention is cheaper than acquisition, so an hour on a past client usually beats an hour on a stranger — and almost nobody spends it that way.
- Loyalty tactics have to match purchase frequency; points schemes fail on people who hire you twice a year.
- Client relationships end over silence, surprises and hidden problems far more often than over bad work.
- Every satisfaction measure is a proxy: NPS and CSAT capture opinion, repeat rate and churn capture behaviour, and comments capture the reason.
- A portfolio piece is evidence when it states a problem, a decision and an outcome — and decoration when it only shows a screenshot.