Ask about the elevator before you promise an arrival window
An arrival time given without knowing the access situation is a promise you will break in front of the customer.
Access decides the job here. A fifth floor walk-up with a narrow stairwell and no elevator is a different crew, a different truck load and a different duration than the same work in an elevator building with a loading area.
Move the access questions to the front of the call. Floor number, elevator or stairs, whether there is a freight elevator, whether it needs to be reserved, and whether a super holds the key.
Booking a two hour window and discovering on arrival that freight is unavailable until Thursday costs you the slot, the drive and the customer's confidence in one go.
The script change is small and pays immediately. Most shops discover they were already asking these questions, just twenty minutes too late in the conversation.
Code every lost job by reason, including the ones the building blocked
Outcome codes are the cheapest instrument you can install, and without them every other decision is guesswork.
Booked. Refused as out of area. No access. Waiting on board approval. Lost on price. Wrong trade. Never reached. Six or seven codes cover almost everything and take a receptionist a few seconds each.
The codes that matter most in New York are the access and approval ones, because they exist nowhere in standard intake software and they usually turn out to be a bigger category than anyone expected.
Once you can see them, the picture changes. Some neighborhoods produce inquiries that are almost never reachable. Some job types die at the paperwork stage rather than at the quote.
Review the codes monthly with the person who answers the phone in the room. The pattern is usually obvious to them long before it is obvious in a report.
The estimate that stalls at the alteration agreement needs its own follow up
Larger jobs here go quiet for reasons that have nothing to do with your price, and a generic follow up sequence treats silence as rejection.
A quote sent to a shareholder in a co-op often has to travel to a board and back. Alteration agreements, insurance certificates naming the right parties, and freight scheduling all sit between agreement and start date. Specifics vary by building and are worth confirming with the managing agent.
Silence during that stretch is not a lost job. Treating it as one, and stopping contact, hands the work to whoever kept in touch.
Build a separate follow up track for approval stage quotes. Longer intervals, useful content rather than pressure, and an offer to supply paperwork directly to the managing agent.
Ask at quote time whether board approval is involved. The answer tells you which track the job belongs on, and it is one question.
A certificate of insurance request should be a same day answer
How fast you can produce paperwork is a conversion factor, not an administrative detail.
Management companies and boards screen vendors on documentation. A firm that takes four days to get a certificate issued with the right parties named loses jobs to a firm that does it in an afternoon.
Find out where the delay actually is. Usually it is your broker's queue, or nobody owning the request internally, and both are fixable without spending anything.
Make it a named responsibility with a target turnaround, and track how often you hit it. Requirements vary by building, so the goal is speed and accuracy in responding, not guessing what will be asked.
Trade buyers notice. A super who got a certificate the same day tends to call again, and that repeat relationship is worth more than most of the marketing spend that produced the first job.
Say the stair carry out loud, before the tech is standing in the lobby
Price surprises at the door are the most expensive conversion failure in this business, and here they usually come from access.
The same water heater, the same repipe, the same drain line costs more in New York because getting materials and equipment to the work is genuinely harder. No driveway, restricted parking, stairs instead of a lift.
Say it during the booking call and say it on the site. Pricing that looks high next to a national average is not high, it reflects a different job.
Shops that explain access costs up front lose a few price shoppers on the phone, which is the point, and stop losing jobs at the door after the drive has already been paid for.
Give the person answering the phone a rehearsed way to say it. Improvising a price conversation is how confident quotes turn into apologetic ones.
Rank the fixes by dollars lost, because your sample will never be big enough
One metro does not produce the volume to prove a small change, so stop pretending a split test is the tool for this job.
Statistical testing needs traffic that most single market trade businesses do not have. Splitting the sample further by borough makes it worse, and a test that never reaches significance produces a conclusion someone will act on anyway.
Rank problems by estimated money lost instead. Unanswered calls in the first hour. Quotes never followed up. Inquiries refused for access after the truck was already assigned. Count them, put a number against each, fix the biggest first.
Some things are genuinely testable because they are large effects. Whether the phone gets answered after hours. Whether an access question is asked at minute two or minute twenty. Those show up without any statistics.
Measure the outcome the same way each month and keep the changes one at a time. Slower than a dashboard full of experiments, and far more likely to be true.
Questions we actually get
- What is the single biggest intake leak for a New York plumbing shop?
- In our experience it is a tie between calls that go unanswered and access questions asked too late. The first is obvious once you count it. The second is invisible until you add outcome codes, and then it usually turns out that a meaningful share of booked jobs were canceled or rescheduled because nobody asked about the elevator, the freight window or the stairwell before promising a time.
- Should we run A/B tests on our booking page?
- Usually not, at least not for small changes. A single metro rarely generates enough qualified traffic to detect a modest effect, and splitting further by borough makes the sample smaller still. Ranking fixes by estimated money lost and making them one at a time is slower to feel scientific and much more likely to produce a real gain. Large effects, like answering the phone after hours, do not need a test to be visible.
- How do we handle quotes that go quiet during board approval?
- Put them on a separate follow up track with longer intervals, and ask at quote time whether approval is involved so you know which track applies. Offer to send insurance documentation to the managing agent directly. Requirements vary by building and are worth confirming with them. The mistake is treating a normal approval delay as a lost sale and stopping contact.
- What outcome codes should we add that most software does not include?
- No access, waiting on board approval, and refused as out of area are the three that matter most here and are usually missing. Add them, use them for a quarter, and the pattern in where your unreachable inquiries come from will be clear enough to change your marketing targeting, not just your intake script.
- Is it worth answering the phone overnight?
- It depends on what share of your revenue is genuinely emergency work, which your own outcome data will show. What is not worth doing is advertising after hours availability and then sending callers to voicemail. If overnight coverage is not realistic, be explicit on the site about when you answer. Missed expectations cost more than an honest limit.
