Why the Second Job Is Harder to Win Than the First
Why this matters
Most shops believe the hard part is getting a stranger to call. It is not. The hard part is getting them to call the second time, and the reason is structural rather than a matter of effort: the first job was sold by the customer's problem, and the second one has to be sold by you. A shop can run a full order book for years while quietly converting fewer than half its first-time customers into anything else, and nothing on the profit and loss statement will flag it, because the revenue arrives either way. The leak only becomes visible when acquisition gets more expensive, which is exactly the moment it is too late to fix cheaply.
The first job is bought under conditions you will never get again
When a stranger calls you the first time, something is broken, leaking, cold, dark, or overrun. That urgency is doing almost all of the selling. It compresses their search, suppresses price comparison, forgives an arrival window you would not accept from anyone else, and makes the decision for them. You were not chosen so much as reached.
The second call is made under the opposite conditions. Usually nothing is on fire. There is time to look around, time to ask a neighbour, and no penalty for delay. The customer now has to decide, in a calm moment, that they prefer you. Preference is a much higher bar than availability, and nothing about a well-executed first job automatically clears it.
This is why the standard advice, "do good work and they will come back," under-delivers. Good work is the entry condition for the second job. It is not the mechanism that produces it.
What actually survives in the customer's head
Six months after a repair, ask a residential customer to name the company that did it. A large share cannot, and the ones who can are usually reading it off something. That is the real state of play, and it means the second job goes to whatever is physically findable inside about two minutes of need:
- A contact saved in their phone, ideally saved by them at the moment of a good experience
- A sticker or tag on the equipment itself, which is the single most durable artifact in the trades
- A paper invoice in a drawer with a folder or a household file behind it
- A magnet, a card in a specific place, or a recurring reminder they set
- Failing all of those, a search box, where you are one result among many and your first job counts for nothing
Note what is not on that list: their memory of how well you did. Satisfaction is what makes them willing to call you again. Findability is what determines whether they can. Shops routinely optimize the first and ignore the second, then read the resulting silence as dissatisfaction.
The default-contractor slot, and how it gets filled
Every household holds roughly one slot per trade: the shop they call without thinking. That slot is not awarded by quality. It is awarded to whoever occupied it at the last moment of need and was easy to reach. A slot filled by a competitor for a trivial reason, a neighbour's offhand recommendation, a fridge magnet from a mailer, is extremely hard to take back later, because taking it back requires the customer to run a comparison they have no motivation to run.
The practical consequence: the contest for the slot is decided in the days right after the first job, not months later when you finally follow up. That is when the invoice is on the counter, the truck is still a recent memory, and saving a contact takes ten seconds. Wait a season and you are not competing for an empty slot, you are trying to evict someone from a full one.
The second job carries the first job's imperfections at full weight
On the first job, small frictions get absorbed. The customer had no comparison, the problem was urgent, and a late arrival or a bill that landed above the verbal estimate registered as "how this goes." Those memories do not disappear; they get filed and repriced.
On the second decision, with no urgency to override them, those same frictions become the whole input. The customer is not weighing your technical work, which they mostly cannot evaluate anyway, against a competitor's. They are weighing one remembered irritation against the effort of finding somebody new, and the effort of finding somebody new is now very low. This is why a shop can have a strong reputation for quality and a poor second-job rate at the same time: the quality was never in question and was never the deciding variable.
The arithmetic of a one-and-done list
All figures below are illustrative but the shape is common. A shop pulls six years of history and counts by unique customer, not by job.
- 640 unique customers were served at least once.
- 260 of those 640 came back for a second job. That is a second-job rate of about 41%.
- 190 of the 260 who reached a second job went on to a third. That is about 73% of the second-job group.
- 155 of the 190 who reached a third went on to a fourth. That is about 82% of the third-job group.
Read those three continuation rates carefully, each against its own immediately preceding group: 41%, then 73%, then 82%. The step from one job to two is by a wide margin the leakiest point in the entire relationship, and every step after it gets easier. Once a customer has chosen you twice, the third choice is close to automatic, because by then a habit exists and the slot is yours.
Now the volume. The 380 one-and-done customers produced 380 jobs between them. The 260 repeaters averaged about 4.1 jobs each, roughly 1,066 jobs. Total across the six years, about 1,446 jobs. So 41% of the customers produced about 74% of the work.
And the leverage. Suppose the shop moves its second-job rate from 41% to 50% of first-time customers. That is 9 points on a base of 640, so about 58 additional customers reaching a second job. At the repeater average of 4.1 jobs each, those 58 produce about 238 jobs, of which 58 were their first jobs and were already in the baseline. The incremental work is about 180 jobs, roughly 12% more total jobs across the same six years, with no additional first-time customers acquired and no additional advertising.
That is the number worth sitting with. A nine-point improvement in one conversion, on customers who already called you once, is worth more than most shops' entire lead-generation spend, and it competes for none of the same budget.
The window where the second job is winnable
The winnable window opens the day the first job closes and narrows fast. Three things are true inside it and false outside it:
- They remember you unprompted. No reminder needed, no explanation of who is calling.
- They have current evidence. The repair is working, the site is clean, the invoice matched the estimate. Every piece of proof you have is at its freshest.
- The slot is empty or weakly held. No competitor has been in the house since.
Every action that converts a first-timer into a second job is cheap inside that window and expensive outside it. Saving your number in their phone, leaving a dated tag on the equipment, agreeing the next service date on the spot, telling them the one other thing you do that they will plausibly need: all of it costs minutes at the tailgate and turns into a phone campaign eighteen months later. The window is not a marketing concept, it is the difference between a two-minute conversation and an hour of chasing.
What shops mistake for loyalty
Warranty visits. A customer who calls you back because your work failed is not a repeat customer; that visit is a cost, and counting it as repeat business inflates your rate and hides the leak. Exclude warranty and callback visits before computing anything.
Convenience. A customer who calls you because you are already saved in their phone from an emergency is holding your slot for a reason that has nothing to do with preference. That is a real asset and worth protecting, but it will not survive one bad experience, and it produces no referrals.
A captive market. In a small town or a specialized trade with two providers, a high repeat rate is geography. It tells you nothing about whether you are winning, and it will read as loyalty right up until a third provider arrives.
Total revenue growth. A shop can grow revenue for years on new customers alone while its second-job rate declines. Revenue is the number that hides this problem; unique-customer conversion is the number that exposes it.
What changes the shape of the problem
Genuinely episodic trades. Where the natural recurrence of the work is measured in many years, a low second-job rate is structural and not a failure. Measure against the asset cycle instead: what share of customers call you when the asset actually comes due, and what share of your work is with people who have used you before. Also look sideways, at whether first-time customers ever buy the adjacent service you offer, because in an episodic trade that is where the second job actually lives.
Commercial and managed accounts. The second job is a procurement decision, not a preference. It is won by being set up correctly as a vendor, by the paperwork being clean, and by being easy for a person who did not hire you originally to keep using. Findability matters less; being on an approved list matters more.
Plan and agreement customers. A maintenance agreement wins the second job at the moment of the first, by contract rather than by memory, which is most of why plans work. That mechanism is a separate subject with its own economics and pitfalls; see the membership references below rather than treating a plan as a substitute for the habits above.
How to tell whether you have this problem
Run one query against your own history, and be strict about the definition. Count unique customers whose first job falls in a window that closed at least two service intervals ago, so recent customers are not counted as failures before they had a chance. Exclude warranty and callback visits. Then compute what share of those customers have a second, self-initiated, billable job.
Take the resulting number, compare it against the 41% in the example above only as a shape rather than a benchmark, and check the trend across three separate cohorts, for example customers first served four, five, and six years ago. The trend is more informative than the level. A rate that is flat and low is a structural issue in how you close a first job. A rate that is falling year over year is usually a specific change you can name: a follow-up routine that quietly stopped, a technician who left, a jump in first-time volume from a channel that sends you people who were only ever going to call once.
References
- U.S. Small Business Administration (SBA), guidance on customer retention economics for small business
- Standard cohort-analysis practice as applied to repeat-purchase measurement
- See related: First-Time to Repeat Customer Conversion; How to Onboard a New Customer Into Your Shop; Customer Lifetime Value (CLTV) for Service Business; Membership + Maintenance Club Programs