Employee Onboarding: Why It Fails and Who Should Own It

Being in Talent Management from 8+ years, I have onboarded more than 250 people a year.

And the failure I have seen most often does not look like a failure at all. The laptop was issued. The handbook was signed. The HRIS record was created on time. Somebody ticked a box that said “onboarding complete” on day two.

Then in week six the same person is still asking a colleague who approves their expenses.

Most organisations think of employee onboarding as a welcome. It is a handover. Somebody knew what this person was hired to do, what they were promised, and what they were nervous about, and that knowledge had to move from one set of hands to another. Every unfinished handover shows up later as a resignation, a missed deadline, or a manager complaining about “fit”.

Here is how badly the handover is going.

Translation: 88% of your workforce is telling you the handover did not happen. Not that it was unpleasant, not that the coffee was bad on day one, but that the transfer of context, access, and expectation they needed to do the job was never completed.

Now put a number on it. An unfilled senior engineering role drains roughly $500 a day in lost output. A hire who arrives on time but takes an extra 40 days to reach full productivity is costing you the same money, quietly, from a seat that finance thinks is filled. That is $20,000 per hire that never appears on a single report.

Here is what this guide contains:

• The onboarding RACI, and why “HR owns onboarding” is the answer that breaks it
• The three failure modes, and a four-line cost model finance can audit on your own numbers
• What to automate, and the four moments you must never automate
• Six practices that survive contact with reality, each with a named failure mode
• The batch model for onboarding 50 or more people at once

HR Insights Lab approach: This page is not a task list. It is the operating manual for the system that produces the task list, written by someone who has run it at volume and watched which parts break first.

The Onboarding Library

This page is the front door. Three companion guides go deeper on the parts that need step-by-step detail:

• The step-by-step guide, stage by stage, for anyone building the sequence from scratch
• The full task list, with owners and completion tests, for anyone auditing what they already run
• How to measure it, with formulas and benchmarks, for anyone defending the budget

What is Employee Onboarding? (And Why it isn’t Orientation)

Employee onboarding is the structured process of integrating a new hire into an organisation’s systems, role expectations, and culture, running from offer acceptance through the first 12 months. It works by sequencing access, knowledge, and relationships so the employee reaches full productivity and decides to stay. Onboarding ends at a measurable outcome, not on a date.

That last sentence is where most definitions collapse. If your definition of onboarding does not contain a number, it is not a definition. It is a to-do list.

Here is what the outcome-based version looks like when you measure it. Across the hiring I have owned, internal movers reached full productivity in 28 days. External hires into comparable roles took 67. Same induction materials, same compliance modules, same handbook. The gap is 39 days of paid output, per hire, and it is entirely explained by how much context arrived with the person.

Here’s the difference:

Four words get used interchangeably in most organisations, usually by the people who control the budget. That confusion is expensive, so define them by owner and exit criterion rather than by activity.

Term	What it is	Who owns it	How long	What “done” looks like
Preboarding	Everything between offer acceptance and the first morning	HR operations	Offer to day one	The hire knows what happens on Monday and their access has been tested
Orientation	The induction event: policies, paperwork, introductions	HR	One to two days	Mandatory paperwork complete and compliance modules assigned
Onboarding	Full integration into role, systems, and relationships	The hiring manager	Offer to month 12	The hire is producing at the expected standard without supervision
Training	Building role capability	The function	Ongoing, no end date	Certification or demonstrated competence against a defined skill

The conflation is not a semantic problem. It is a funding problem. An organisation that calls orientation “onboarding” declares victory on day two and stops investing, because by its own definition the job is finished.

The Four C’s, and The Fifth one Everyone Skips

Dr Talya Bauer’s four C’s framework, developed for the SHRM Foundation, is the most durable model in this space and worth keeping. What follows is not the four bullets you have already read. Each C is listed with the failure mode that appears when it gets skipped, because that is the part nobody publishes.

• Compliance. Rules, paperwork, mandatory training. Skipped, it becomes a legal exposure. It is also the only C that is fully auditable, which is exactly why it is the one that never gets skipped and the one everyone over-indexes on.
• Clarification. What the role is and what good performance looks like. Skipped, the hire spends month one guessing, and their first deliverable is judged against a standard nobody told them about.
• Culture. The unwritten norms: how decisions get made, who to ask, what gets escalated. Skipped, the hire commits an avoidable political error in week three and quietly loses credibility they never recover.
• Connection. The relationships that make the work possible. Skipped, the hire has a manager and nobody else, so every question costs them a favour and they stop asking.
• Capability. The fifth one. Can this person do the work with the tools and access they have been given? Skipped, everything above is theatre.

Capability fails silently, and it fails for a structural reason. Compliance is auditable. Culture is felt. But nobody checks on day 40 whether the new hire has write access to the system they were hired to run.

I hire on capability signals over credentials. It would be absurd to then onboard on paperwork signals.

Why Onboarding Fails: The Real Cost of a Broken First 90 Days

Nearly every guide on this topic moves from “what onboarding is” straight to “here are the steps”. That sequence has a hole in it. If you do not know why the last process failed, the new one fails the same way with better formatting.

Here’s what breaks:

Failure Mode 1: The Orphaned Handover

Recruitment owns the candidate until the offer letter. The manager owns them from around day 30, once the noise has settled. Nobody owns the 30 days in between.

You already know the symptoms. No manager one-to-one scheduled in week one. An equipment request raised and never chased because the person who raised it assumed the ticket was the job. Six weeks of project context that lived only in the recruiter’s head, and the recruiter has moved on to the next requisition.

Why this happens:

Onboarding is the only significant HR process with no single accountable owner. Payroll has one. Performance review has one. Onboarding has five contributors and an assumption.

My own work on the sourcing side makes the waste obvious. When I run talent mapping across a competitor set, I build a detailed picture of each candidate months before a job description exists: what they are working on, what would move them, what they are trying to get away from. That intelligence survives every stage of the hiring process. Then it evaporates at the offer letter, because the artefact that carries it does not exist.

Failure Mode 2: Onboarding That Ends on Day Two

The organisation runs a genuinely good day-one experience. Welcome pack, leadership video, a properly produced induction. Satisfaction scores come back high, HR reports them upward, and everyone concludes the programme works.

Then week three arrives and the new hire has no structured contact with anyone.

The measurement is the problem. Most onboarding surveys are fielded at day seven, which is precisely the window where the honeymoon effect is strongest and the data is least useful. You are surveying someone who is still relieved to have got the job, and reporting their relief as programme quality.

Which means the damage happens in a window nobody is measuring.

Deloitte finds that one-third of new hires are likely to leave within their first six months, and that workers who felt their onboarding was highly effective are 18 times more likely to feel highly committed to their organisation. Translation: a third of your intake is already leaving, the decision is forming in months two to five, and your survey closed in week one.

Source: Deloitte, New Hire Onboarding

Early attrition then gets attributed to pay, or to “fit”. It is almost never either. Retention problems are mostly systems failures, not compensation failures, and the first 90 days is where the system either transfers what the person needs or does not.

Failure Mode 3: Process Without Capability

The new hire completes every module and still cannot work. Compliance training done. Handbook signed. Culture deck watched. No access to the production environment, no introduction to the three people who approve their output, no clarity on what their first deliverable is or when it is due.

Administrative completion is not operational readiness. Most onboarding dashboards track only the first one, because it is the only part that is easy to instrument.

Here’s the difference:

One of those is a compliance metric. The other is a business metric. Only one of them has ever won a budget argument.

Go back to the 28 days versus 67 days split. Internal movers hit productivity in roughly 40 to 50% of the time it takes an external hire, and they do not receive a better induction. They receive nothing extra at all. What they already have is the context: the systems, the people, the politics, the unwritten rules about how work gets approved. That isolates context transfer as the active ingredient, and it means the 39-day gap is not a fact of nature. It is a design choice.

The Four-Line Cost Model

Finance does not fund engagement scores. It funds models it can audit. Here is one you can build on your own cost base in an afternoon, with four lines and no assumptions borrowed from someone else’s benchmark report.

Line	How to calculate it	Worked example: one senior engineering hire
1. Extended ramp cost	Extra days to productivity x loaded daily cost	39 days x $500 = $19,500
2. Manager drag	Hours re-explaining context that should have transferred x manager loaded rate	3 hrs/week x 8 weeks x $95 = $2,280
3. Replacement cost if they leave inside 12 months	Full cost per hire, paid a second time	Your cost-per-hire figure, in full
4. Opportunity cost during the re-hire	Vacancy days x loaded daily cost	60 days x $500 = $30,000

Lines three and four are the ones that change the conversation, because they are the same money twice. On the senior engineering roles I have owned, cost per hire ran between ₹6.5L and ₹7.5L before we rebuilt the sourcing model and between ₹3L and ₹3.8L after, a reduction of 45 to 55%. That saving is real and it is also fragile. It exists only until a hire made at the lower number walks out at month nine and you pay the original number again.

Two notes on using this. Run it on one real role rather than an average, because averages invite arguments about whether the average applies. And treat it as a planning tool rather than a financial projection, which is also how you should present it.

What you will not find in this model is the “it costs six to nine months of salary to replace an employee” figure that appears on almost every competing page. It is untraceable, and the first thing a competent finance partner will ask is where it came from. Build from cost per hire you can evidence instead.

The Five Stages, End to End

Five stages carry a new hire from offer to full contribution. Preboarding runs from acceptance to the first morning. Day One is its own stage because it sets the tone for everything after it. The First Week establishes working rhythm. The First 90 Days builds capability. Months Four to Twelve is where the second attrition wave lives.

The exit criteria:

The stages are not the interesting part. What has to be true before each one closes is.

• Preboarding closes when the hire knows what happens on Monday and their access has been tested by a human, not just provisioned.
• Day One closes when they have had a real conversation with their manager about the work, not the org chart.
• The First Week closes when they know their first deliverable, its date, and who receives it.
• The First 90 Days closes when they have shipped something a colleague depended on, unsupervised.
• Months Four to Twelve closes when they are performing at the standard the role was scoped for and there is a next step they can name.

Stage names are where most guides stop. Exit criteria are where the system starts. The sequencing detail, what happens inside each stage and in what order, is where the execution lives, and that is covered separately in the complete employee onboarding process, stage by stage.

Who Owns Onboarding (and Why “HR” is the Wrong Answer)

Ask ten organisations who owns onboarding and nine will say HR. That answer is the reason it fails.

HR cannot be accountable for whether a person becomes productive in a job HR does not do, on a team HR does not sit in, using systems HR does not administer. What HR can be accountable for is that the system runs. Those are different things and confusing them is why HR gets blamed for outcomes it has no lever on.

How to execute:

Build a RACI across the five stages. Responsible does the work. Accountable carries the consequence. Consulted is asked before. Informed is told after. This is the version I run.

Stage	HR Ops	Recruiter	Hiring Manager	Buddy	IT	New Hire
Preboarding	A	R	C	I	R	I
Day One	R	I	A	R	C	I
First Week	C	I	A	R	I	R
First 90 Days	C	I	A	C	I	R
Months 4 to 12	A	I	R	I	I	R

One rule holds the whole table up: exactly one Accountable per stage. Never two. The moment a stage has two Accountables it has none, because each will assume the other has it.

The manager is Accountable for three of the five. That is deliberate, and it is the part that gets pushed back on.

When a manager says onboarding is HR’s job, the counter is not a values argument. It is this: you were accountable for the hiring decision, you will be accountable for the performance review, and you are the only person who can tell whether this hire is capable yet. HR cannot hold accountability for a judgment only you are qualified to make. What HR will do is remove every piece of administrative work between you and that judgment.

Where the table flexes: the Consulted and Informed columns move freely by organisation size. In a 200-person business the buddy and the manager are often the same person and that is fine. Where it does not flex: one Accountable per stage, and the manager holds Day One through the first 90 days. Everything else is negotiable.

This is the Talent Supply Chain applied past the point most organisations stop applying it. Treating hiring as a supply chain means every handoff has an owner and a spec. Onboarding is the final leg of that chain, and it is the leg most organisations leave unowned.

The Context Transfer Document

The RACI names who is accountable. This is the artefact that makes the first handoff survivable.

The CV tells the manager what the candidate has done. The context document tells them who they are getting. Six fields, and none of them optional:

• What the candidate was told about the role, in the words that were used
• What they asked about repeatedly across the process, which is what they care about
• What they are leaving behind, and why they are leaving it
• What they said they want to work on next
• What they were nervous about
• Any commitment made verbally during the process that is not in the offer letter

Field six is the one that saves you. Verbal commitments made in a final-stage conversation, about scope, about a project, about a title in 12 months, are the single most common source of the week-three moment where a new hire realises the job is not what they were sold.

Keep it role-relevant only. No health information, no family circumstances, no protected characteristics. It contains personal data gathered during hiring and should be retained under the same policy and the same retention period as your other recruitment records.

My intelligence-led sourcing work generates exactly this material as a by-product. Mapping a competitor set produces detailed intelligence on dozens of people. The waste is not in gathering it. The waste is throwing it away at the offer stage.

If frameworks like this one are useful to you, the Lab newsletter sends the next one before it goes anywhere else.

The Manager’s Role: The Single Biggest Variable

Onboarding quality is largely a function of one person’s calendar.

Not the induction deck, not the platform, not the welcome pack. Whether the hiring manager blocked the time. Every other input is a rounding error next to this one, and most guides list it as one bullet among ten.

Here’s the deal:

Active involvement is a specific set of behaviours, not an attitude. Present on day one, in person or on video, for a conversation about the work. Weekly one-to-ones from week one, not from month two. Owning the plan rather than receiving it from HR. Running the check-in gates personally. Making the introductions themselves rather than sending a list of names.

Here is what that involvement is worth.

Translation: seven in ten new hires start the job knowing they are not ready for it. They will not say so, because week four is a bad time to admit you are lost.

Now the harder half. Managers do not skip this because they do not care. They skip it because no time was allocated, because onboarding outcomes appear nowhere in their own goals, and because nobody ever showed them what good looks like. Given a busy week and no forcing function, the new hire loses to whatever is on fire.

Managers do what they are measured on. If onboarding is not measured, it is optional, and optional loses every time. The manager’s calendar is the system, and a system with no forcing function is a suggestion.

The Manager Enablement Pack

“Be involved” is an expectation. A pack is a tool. Hand every manager these five artefacts before the start date, never on day two, because a pack delivered after the hire arrives is a document nobody reads.

• A pre-populated plan template for that specific role, not a blank one
• The context transfer document from the recruiter
• Three suggested week-one deliverables appropriate to the role
• Check-in agendas, each naming the decision that check-in has to produce
• The buddy assignment, already made and already accepted

Do not train the behaviour. Engineer it. Training is what organisations do instead of building the tool, and a workshop attended in March does not survive a busy week in September.

The drafting work in that pack, the role-clarity translation, the first-pass agendas, the plan skeleton, is exactly the kind of task that AI as a recruiting co-pilot should absorb, so the manager’s time goes to the conversation rather than the document. More on where that line sits below.

Six Onboarding Practices That Survive Contact with Reality

Most best-practice lists give you ten items of equal weight, no difficulty rating, no cost, and no indication of what happens when you implement one badly. That is why you have read four of them and implemented none.

Advice without a failure mode is advertising. Every practice below names the specific way it breaks.

Start Onboarding at Offer Acceptance

How to execute:

Treat the acceptance email as the start signal, not the finish line. Access requests raised that day. The context transfer document written inside 48 hours. The manager makes contact within the first week of the notice period, about the work, not about logistics.

Why this works:

The notice period is the only free time you will ever get. The hire is committed, has capacity to absorb context, and is at their most anxious about the decision. Silence during that window is what a counter-offer is built on.

Appoint Buddies on a Power Law

How to execute:

Stop assigning buddies on rotation. Identify the 5 to 10% of people who are genuinely good at it, ask them, give the role a name and visible status, and use them repeatedly. Accept that the distribution will be uneven and defend it when someone calls it unfair.

Why this works:

Helping a new person integrate is a skill, and skills are not evenly distributed. A rota optimises for fairness among existing employees. It does not optimise for the new hire, who is the person the scheme exists to serve.

Put 90-day Productivity in Manager Goals

How to execute:

Write one line into the goal sheet of every manager who will hire this year. The measure is the date of the new hire’s first unsupervised deliverable, not a satisfaction score. The reviewer is the manager’s own manager, in the normal performance conversation, not HR in a separate one.

Why this works:

This is the practice that makes the other five stick. Everything else on this list is an improvement to a process that remains optional. This one changes who carries the consequence, which is the only thing that reliably changes behaviour at scale.

Write Goals a Third party can Verify

How to execute:

Apply one test to every onboarding goal you write: could someone outside the relationship confirm whether it happened? “Understands the product roadmap” fails. “Has presented the roadmap back to two stakeholders and answered their questions” passes. The stage-by-stage guide covers how to build the plan itself; this is the rule for what goes in it.

Why this works:

Definitional specificity is what produces measurable improvement, and I have watched this hold on the hiring side repeatedly. Running competition analysis before opening a role lifted Quality of Hire by 12%. Mapping the talent landscape across similar industries added 14%. Building skill-based pipelines off proper market analysis produced a 30% improvement. In each case the intervention was the same: define precisely what you are looking for before you go looking. Vague onboarding goals fail for the same reason vague role definitions do.

Survey After The Honeymoon, Not During

How to execute:

Move the primary survey window out of the first fortnight. Keep a light day-one pulse if you want an operational check on logistics, but stop treating it as a measure of onboarding quality. The score that matters is collected once the novelty has worn off and the person has had time to discover what they were not told.

Why this works:

A new hire in week one is answering a different question from the one you asked. They are telling you how they feel about having been hired. Later, they tell you whether the handover worked. Only the second one is actionable, and only the second one predicts whether they stay.

Design for The Second Attrition Wave

How to execute:

Extend the process past 90 days with a small number of deliberate touchpoints in months four to twelve. Not more induction. A conversation about what the role has turned out to be, a check on whether the growth they were promised is visible yet, and a named next step they can work towards.

Why this works:

The first attrition wave is a mis-hire or a broken start. The second one is different. It happens once the person is competent, has stopped being interesting to their manager, and has enough distance to compare what they were sold against what they got. That comparison happens around month seven, and by then almost every organisation has closed the file.

Digital Employee Onboarding: What to Automate and What to Never Automate

Almost everything written about digital employee onboarding is written by someone selling onboarding software. That does not make it wrong. It makes it incomplete in one predictable direction: every article concludes you should automate more, because every author benefits if you do.

I have nothing to sell you. So here is the neutral version.

Here’s the rule:

Automate anything that is high-volume, low-judgment, and identical across hires. Three conditions, all of them required. A platform’s feature list is not a strategy, and “our tool supports it” is not a reason to automate something.

What to automate

• Document collection and e-signature
• Compliance and mandatory training assignment, plus completion tracking
• Equipment and access provisioning, triggered by the offer acceptance event rather than by a human remembering
• The preboarding communication sequence across the notice period
• Check-in reminders to managers
• Survey scheduling
• The task tracker spanning HR, IT, facilities, and the manager

Two of those deserve a caveat. Verification requirements for employment eligibility vary by jurisdiction, and e-signature is not universally accepted for right-to-work or I-9 documentation. Check the rule that applies where the person will work before you automate that step.

What to never automate

Here’s what breaks:

Four moments lose all their value the instant they are automated. Not most of their value. All of it.

The manager’s day-one conversation. Replaced by a recorded welcome, it communicates the opposite of what it intends: that this hire was not worth 30 minutes of a real person’s time on the one day they will remember.

The buddy relationship. An automated match produces a name in a field. Nobody has been asked, nobody has agreed, and the new hire is now expected to initiate a relationship with a stranger who does not know they were assigned.

The gate decisions. Whether this hire is on track is a judgment about capability in context. A system can tell you a milestone was missed. It cannot tell you whether that matters.

The “how are you finding it” conversation at week three. Replace it with a survey and you collect a score and lose the reason. The reason is the entire value, and it only ever arrives unprompted, in a conversation, from someone who trusts the person asking.

Keeping those four human costs manager time, and that cost has to be planned for rather than wished away. Roughly four hours across the first 90 days, per hire. Budget it or it will not happen.

Where AI is Useful in Onboarding

Point AI in recruitment at the repetitive-question layer and the stall-detection layer, keep humans on judgment, and the economics work. The broader argument for human and AI teaming in HR, and where agentic AI in HR is heading next, is worth a longer read on its own.

Onboarding at Volume: The Batch Model for 50+ Hires

Every guide on this topic assumes you are onboarding one person. Run that same process fifty times in parallel and it collapses in week two, usually at the point where one HR coordinator is fielding forty variations of the same question.

Here’s what changes:

• Cohort starts, not rolling ones. Fixed start dates, so orientation runs once for thirty people rather than thirty times for one. This is the single change that makes everything else possible.
• Batch orientation with role-specific breakouts. The shared 60% runs collectively. The role-specific 40% splits by function in the afternoon.
• Cohort buddies, not one-to-one. At volume the buddy ratio breaks first. One strong buddy to five or six new joiners in the same function works. One to ten does not, and one to one is arithmetic you cannot fund.
• Standardised plan templates per role cluster. Not per role, per cluster. Fifteen templates, not four hundred.
• A daily stand-up during ramp weeks. Fifteen minutes, HR and the managers, exceptions only.
• One tracker, visible to HR, managers, and IT simultaneously. The moment there are two trackers, there are none.
• A single named point of contact per batch. Forty people with one number to call is manageable. Forty people escalating to whoever answers is not.

The design work behind this belongs to the same discipline as bulk hiring and ramp-up planning, and if you are onboarding at this scale you are almost certainly hiring at it too.

What to Track, and What to Ignore

Four metrics. Not twelve. Each one earns its place by answering a business question rather than describing an activity.

Metric	The question it answers	My benchmark
Time-to-Productivity	How long are we paying for reduced output?	28 days internal, 67 days external
90-day and 12-month retention, split by manager and hiring source	Where is the leak?	89% at 12 months
Onboarding NPS, fielded after the honeymoon	Is the experience working once novelty ends?	Compare against your day-7 score
Manager onboarding compliance	Is the system being run at all?	Track it before you improve it

Here’s the one that matters:

The fourth is the only leading indicator. The other three report what already happened, which makes them useful for diagnosis and useless for intervention. Manager compliance tells you today whether next quarter’s retention number is being built or broken.

Measuring it is worth the effort because the upside is large. Gartner finds successful onboarding can increase discretionary effort by more than 20% and drive performance by up to 15%. Translation: this is a productivity lever, not a hygiene factor, and it should be argued for as one.

Source: Gartner, Onboarding New Hires Checklist for HR Leaders

Naming the four is the strategy. The formulas, collection methods, benchmarks, and survey question sets are a separate job, and that job is measuring employee onboarding.

What good looks like, task by task

You have downloaded four onboarding checklists and used none of them. The reason is structural, not motivational.

Almost every checklist in circulation has two columns: phase and task. A task list with no owner is the orphaned handover from earlier in this article, rendered in bullet form. It looks like a control. It is a reminder.

Here’s the difference:

A working checklist needs four columns. Phase, Task, Owner, and Done when. Owner maps directly to the RACI. Done when has to be verifiable by a third party, which is where most checklists quietly fail.

Phase	Task	Owner	Done when
Preboarding	Provision system access	IT, verified by manager	New hire has logged into all four core systems and completed one transaction in each
Week one	Agree the first deliverable	Hiring manager	Deliverable, due date, and receiving colleague are written down and shared with that colleague
Month six	Growth conversation	Hiring manager	A named next step exists that the employee can repeat back unprompted

Look at row one. “IT access confirmed” is what most checklists say, and it is wrong, because it can be true while the person still cannot work. The version above cannot be ticked by someone who has not checked.

Those three rows are a demonstration of the structure, not the artefact. The full version, phase by phase across the whole year, is the 12-month employee onboarding checklist.

Frequently Asked Questions

The Last Leg of the Chain

Onboarding is not a welcome programme. It is the last leg of the Talent Supply Chain, and it is the leg most organisations leave unowned.

The gap between organisations that get this right and organisations that keep rebuilding the same broken process is not budget, and it is not software. It is whether one named person is accountable for whether the new hire can do the job by day 90.

That question is about to get harder to avoid. Gartner expects HR to turn roughly a third of its recruiting capacity inward in 2026, and regrettable retention to become the primary productivity barrier. If a third of your hiring becomes internal movement, onboarding stops being a new-hire process and becomes an internal mobility process too. Every one of those moves needs the same context transfer and the same single accountable owner, at a point where the business has already decided it cannot afford to lose the person.

Source: Gartner, Four Trends Talent Management Leaders Should Prepare for in 2026

So pick two. Not six.

One from ownership: the RACI, or the context transfer document. One from the practice list, whichever difficulty rating matches the capacity you have this quarter. Run them for 90 days. Measure Time-to-Productivity before you start and again at the end, in days, and convert the difference to currency before you show anyone.

That number is the argument. Everything else is a conversation about how nice the welcome pack looked.

If you are rebuilding onboarding in your own organisation this year, I would like to hear how it goes. Connect with me on LinkedIn and tell me which two you picked.