Employee Onboarding Best Practices When You Deliver Your Own Orders

Most employee onboarding best practices published online were written for a salaried office hire with a laptop, a desk and a manager two metres away. Almost none of that survives contact with a business where the new person spends day three alone in a van, forty minutes from anyone who could answer a question.

The gap shows up in the numbers. Around 20% of employee turnover happens within the first 45 days, and roughly one in three new hires leave inside 90 days, according to SHRM data on early attrition. For an eleven-person operation with three drivers, that is not a rounding error in a workforce plan. It is a route you are covering yourself on a Saturday.

This piece is about quality rather than sequence: which practices change whether someone is still there in month four, which ones are decoration, and what specifically goes wrong in a business where new hires work alone early. If you need the order of operations instead, that lives in the step-by-step employee onboarding process from offer to day 90.

The Bottom Line

• Early turnover is concentrated, not spread. Roughly 20% of it lands in the first 45 days, which is where attention should go.
• The practices that move retention are unglamorous: a named buddy, working logins on day one, expectations in numbers, and a real check-in at day 30.
• Only 12% of employees strongly agree their employer onboards well (Gallup). Doing the basics consistently puts you ahead of most employers.
• Delivery roles have a specific failure mode: the new hire is alone, early, with no one to ask. Design for that rather than around it.
• Culture decks, welcome swag and engagement surveys are not what a six-person operation is short of.

Why early turnover clusters in the first six weeks

People do not usually quit a new job because the work was hard. They quit because the first weeks told them something: that nobody was expecting them, that the job was not what was described, that asking questions was unwelcome, or that they were going to fail publicly and soon.

Each of those is a signal sent during onboarding, and each is avoidable without spending money. That is what makes early attrition worth working on: the causes are cheap to fix and the alternative is expensive. Replacing a driver means advertising, interviewing, licence checks, and three to six weeks of covering the route in the meantime.

The research points the same way. Gallup reports that only 12% of employees strongly agree their organisation does a great job of onboarding new people, and just 29% say they feel fully prepared and supported after it. The bar is low. Consistency, not sophistication, is what clears it.

The seven onboarding practices that change retention

These are the ones that survive contact with a small operation. None require a budget.

1. Name a buddy, and tell the buddy. One experienced person, assigned to one new hire, for the first fortnight. Not “ask anyone”, which means nobody. The buddy’s job is to be interruptible, which means telling them they are the buddy and accepting that their own output dips slightly for two weeks. An unnamed buddy is a hope, not a practice.

2. Do the paperwork before they arrive. Every form completed in advance is an hour of day one returned to actual work. A new hire who spends their first morning filling in tax forms has learned that the job is administrative and that you were not ready. Both are wrong impressions and both are self-inflicted. The full list of what to collect, issue and sign before a first shift is the thing to work through the week before, not on the morning.

3. Test every login while they are standing next to you. Scanner, route app, timekeeping, fuel card, building access. Each one, actually used, on day one. A login that fails on day three, out on a route, with nobody answering the phone, does more damage than a week of bad training. This is the single most common avoidable failure in delivery onboarding.

4. Put expectations in numbers. “Twelve to sixteen drops a day by the end of month two” can be met, missed or queried. “Getting up to speed” cannot. People who do not know what good looks like assume they are failing, and people who assume they are failing start looking elsewhere before anyone tells them otherwise.

5. Escalate difficulty deliberately across the first week. Ride along, then handle the stops, then drive with the buddy present, then run a short route solo. Each day adds one thing. Sending someone out alone on day two because you are short-staffed is the decision that costs you the hire.

6. Book the day 30 check-in before they start. Put it in the calendar during preboarding. Month one is when attention naturally drifts back to operations, which is exactly when the new hire is deciding whether this is working. A calendar entry beats an intention.

7. Ask what was different from what they expected. At day 30 and again at day 90. The answer audits your job ad and your interview, not the employee. If three hires in a row say the early starts were not clear, you have a hiring problem producing what looks like an onboarding problem.

What to do differently when a new hire works alone

This is where generic advice stops being useful. An office hire who is confused turns around and asks. A driver who is confused is alone, behind schedule, standing at a locked delivery door.

Four adjustments:

• Give them a script for failure, not just for success. What to do when nobody answers, when a chiller reads wrong, when an account refuses a delivery, when the van will not start. Write down who to call in each case. Most new-hire panic is procedural, not technical, and a laminated card in the cab solves most of it.
• Make the first solo route an easy one. Familiar accounts, a reasonable drop count, no loading dock puzzles. The first solo run should build confidence, not test it. Save the awkward accounts for week three when a bad day is recoverable.
• Establish that calling is correct behaviour. Say it explicitly: in your first month, call me rather than guess. Then answer warmly when they do, including the fourth time. If the first call is met with audible impatience, there will not be a fifth, and the mistakes will arrive silently instead.
• Debrief at the end of each of the first five days. Ten minutes when they get back. What went wrong, what was confusing, what took longer than expected. This is where you learn that your route notes are out of date and that two customers have moved their delivery window.

Four onboarding practices that waste your time at this size

Not everything recommended in the category earns its place in an eleven-person business.

• The culture deck. Values slides are written for companies large enough that culture has to be described rather than observed. In a small operation the new hire learns your culture from how you speak to a customer who is complaining. The deck adds nothing.
• Welcome swag. A branded water bottle is pleasant and has no measurable relationship to whether somebody is still employed in month four. Spend the equivalent on better safety boots.
• The 30-page handbook read on day one. You need the handbook to exist and to be signed. You do not need it read in one sitting, and nobody does. Preboard it, flag the five clauses that carry consequences, and move on.
• Engagement surveys in month one. A pulse survey asks a new hire to evaluate a workplace they have not yet understood, and it substitutes a form for a conversation you could simply have. At this size, ask them directly.

The general rule: if a practice was designed to scale a manager’s attention across many people, and you have one manager and eleven people, you are better off spending the attention directly.

How to tell whether your onboarding is actually working

Three signals, none of which require a survey tool.

• Survival at day 90. Count how many of your last ten hires reached three months. If it is under seven, the problem is early and structural rather than about who you hired.
• Time to independent work. How many days until a new driver runs a full route unsupervised at acceptable pace? Track it across hires. If it is climbing, your process is degrading or your routes are getting harder without the training changing.
• Volume and timing of questions. A new hire who asks a lot in week one and little by week four is onboarding correctly. One who asks nothing in week one is not confident; they are hiding, and you will find out in week six.

Retention at 90 days is the number worth writing on a wall. Everything else is a proxy for it.

What onboarding software helps with, and what it cannot fix

Software removes administrative failure. It collects documents before day one, chases signatures, tracks licence expiry, and fires the task reminders that stop equipment being forgotten. Those are real gains and they eliminate several of the bad first impressions described above.

What it does not do is supply a named buddy, an escalating first week, a manager who answers the phone patiently on the fourth call, or a check-in at day 30. Those are the practices retention actually rests on, and they are all free. If you are weighing up tools anyway, what onboarding software does for a small delivery team and when it is premature covers the buying decision. Buy it to automate a process that works, not to invent one.

Frequently asked questions

What are the most important employee onboarding best practices for a small team?

Assign one named buddy for the first fortnight, complete paperwork before day one, verify every login and access while the new hire is with you, state performance expectations as numbers, escalate difficulty across the first week rather than all at once, and hold a real conversation at day 30. Those six cover most of the avoidable early turnover.

How long should onboarding last for a driver or warehouse hire?

Ninety days. Early attrition concentrates in the first six weeks and continues through month three, so ending the process at day seven closes the structure exactly when the risk begins. The later stages can be light, a widened rota and two conversations, but the period should stay open.

What is the biggest onboarding mistake small businesses make?

Sending someone out alone too early because the rota is short. It happens for understandable reasons and it reliably produces a bad first week, an avoidable mistake at a customer site, and a hire who concludes they were set up to fail.

Should new hires get a written onboarding plan?

Yes, and it can be one page. What happens before day one, what day one looks like, what each day of week one contains, and the dates of the day 30, 60 and 90 check-ins. Giving that page to the new hire matters as much as having it, because it tells them somebody planned for their arrival.

How do I onboard well when I am short-staffed and have no time?

Protect two things and let the rest slide: paperwork completed before day one, and a named buddy for the first week. Preboarding costs about an hour and removes most day-one friction. The buddy costs half an hour a day from one existing person. Everything else can be added later once the hire is stable.

Where to start

Pick the last person who left inside three months and work out which week they decided to go. Usually it is identifiable, and usually it traces to one of four things: nobody was expecting them, the logins did not work, they were sent out alone too early, or nobody asked how it was going until it was too late.

Fix that one first. Then add the day 30 check-in to your calendar for whoever you hire next, because it is the practice most often intended and least often done.

None of this is sophisticated. That is rather the point. The bar in the wider market is low enough that doing ordinary things reliably is an advantage in keeping people.