Corporate AI7 min read

Which Staff Should We Train on AI First?

On this page

Train leaders and line managers first, because they set the rules and decide whether staff get time to practise. Next, train the roles that spend most of the day on documents, email and Excel, since that is where results show fastest, and pick your champions from them. Customer-facing and frontline teams follow once the data rules are clear. Everyone should hear the basic data rules early.

TL;DR:

  • Leaders first, briefly: they decide direction, budget and risk.
  • Line managers next: without them, training fades within weeks.
  • Then heavy document, email and Excel roles, where time savings are easiest to see.
  • Name two or three champions from those early groups.
  • Customer-facing and frontline staff follow, with short, practical sessions and firm data rules.

Who this guide is for

This is for the HR, L&D or operations lead who cannot train everyone at once. For scheduling a large group once you know the order, see AI training for a 200-person team.

Why the order matters

Staff copy what leaders and managers do. If juniors learn AI and their managers do not, managers distrust AI-assisted work, staff hide their use, or practice time gets cancelled for "urgent" work.

Early groups also find the useful tasks and the gaps in your rules, so later groups get better training.

The five groups, in order

Order Group Why at this point What they need most
1 Executives and senior leaders They set direction, budget and risk A short session on decisions, risk and policy
2 Line managers and supervisors They give staff time and review AI-assisted work How to check AI output and set team rules
3 Heavy document, email and Excel roles Results show fastest and are easy to measure Hands-on practice on their own tasks
4 Customer-facing teams High value, but customer data raises the risk Firm data rules, tone and when to hand over
5 Frontline and shift staff Short, practical uses once the rules are tested Short sessions, phones, plain language

1. Executives and senior leaders

Leaders do not need a long course. They need enough hands-on time to judge proposals, set a risk position and back the policy. A short private session works. For boards, see board-level AI orientation.

2. Line managers and supervisors

Managers are easy to skip, yet they often decide whether training sticks. Managers answer three questions for their teams: what may we use it for, what must never go in, and who checks the result. See getting middle managers to use AI.

3. Heavy document, email and Excel roles

Finance, HR, procurement, executive assistants, communications, monitoring and evaluation, and anyone who writes reports. Their tasks repeat, so you can time them before and after. Early wins here give you evidence for the rest of the rollout.

Champions, picked from the early groups

Champions are not a separate wave. Pick two or three people from the first groups who produce good AI-assisted work and explain it well. Get their managers to agree protected hours. See our champions programme guide.

4. Customer-facing teams

Customer care, sales and front-office staff can gain a lot: faster replies, better summaries, help moving between English and Kiswahili. They also handle the most personal data, so train them once earlier groups have tested your data rules. See training customer service teams.

5. Frontline and shift staff

Drivers, warehouse, ward, shop-floor and hotel teams need short sessions, phone-first tools, plain language and a few clear rules. They benefit most once their supervisors are trained and can support them. See how to train shift workers and frontline staff on AI.

Everyone gets the basic rules early

The order is for the deeper training. The basic data rule should reach everyone in the first weeks, because some staff may already use free AI tools on their phones. A short message or a ten-minute briefing is enough: which tools are allowed, what must never go in, and who to ask.

Kenya's Data Protection Act, 2019 says personal data must be limited to what is necessary for its purpose (section 25(d)). Customer names, ID numbers, phone numbers and M-PESA codes do not belong in a public AI tool.

A simple way to rank your teams

With many departments, score each team from 1 to 3 on these five points. After leaders and managers, start with the highest totals.

Question Score 3 if
How much of the day is reading, writing or spreadsheets? Most of the day
How repetitive are the tasks? The same tasks every week
How easy is it to keep personal data out? Easy, the work uses little personal data
Is the team's manager supportive? Yes, and will give practice time
Will results be visible to leadership? Yes, the outputs reach senior people

This scoring is our suggestion. Adjust it to your organisation.

A worked example

This is a worked example with made-up numbers, not a real client.

A company of 120 staff plans three waves over about three months.

  • Wave 1, month 1: 8 senior leaders in one short session, then 15 line managers in one group. Everyone receives the one-page data rule.
  • Wave 2, month 2: finance, HR and procurement (about 40 people) in two groups. Two champions are named from these groups.
  • Wave 3, month 3: customer care and sales (about 35 people) in two groups, then 20 warehouse and delivery staff in short sessions timed around their shifts.

By wave 3, the rules and examples have been tested by earlier groups.

Mistakes to avoid

  • Training only the enthusiasts. The rest of the team never catches up.
  • Starting with IT alone. IT matters for tools and security, but most gains sit in everyday office work.
  • Training everyone at once. You lose the lessons from early groups.
  • Skipping managers. Staff then get no time to practise.
  • Putting customer-facing staff first without rules. That is where a data mistake costs most.

How FuKazee helps you decide who goes first

All facts below come from our live pages, checked on 15 November 2026.

  • We frame AI around each role: C-suite and board, department heads, operations and admin, finance and risk, marketing and comms, IT, legal and compliance, and HR and learning (AI training in Kenya).
  • Some organisations start with awareness training for everyone, then go deeper with functions such as HR, operations, finance, communications or IT.
  • Leaders can have a private session. Our AI training for executives runs for a CEO, a senior team or a board, led by our founder, Rodney Maina.
  • Managers have their own sessions. Our AI training for managers is for supervisors, branch heads and team leaders, and we can stagger groups so a branch or shift stays covered.
  • We leave champions behind. The people who take to it fastest are visible by the end of module three (in-house vs external).

We believe this order, leaders and managers first with champions from the early groups, is the best way for a Kenyan organisation to start. Our programme supports it: separate sessions for leaders and for managers, and champions named during the programme. FuKazee has trained more than 3,000 people since 2022, and Rodney leads the corporate sessions himself (trainer profile). We are rated 4.9 from 38 Google reviews, as shown on our reviews page on 15 November 2026.

When FuKazee may not be the right fit

  • You want every team trained in the same week. Our founder leads sessions himself, so large rollouts run group by group.
  • You need NITA-accredited training. FuKazee is not yet NITA accredited. Ask us for the current status.
  • You want a self-paced platform for all staff. Our training is taught live, on-site or online.

Questions people ask

Should we train the people who are most excited about AI first?

Include them, but do not stop there. Enthusiasts often make good champions, but a rollout built only on them leaves the rest behind.

How soon should frontline staff be trained?

They should hear the basic data rules in the first weeks. Deeper training works best once their supervisors are trained and can support them.

Do executives really need hands-on training?

Yes, briefly. A short session on real decisions helps leaders judge the proposals and budgets they approve.

How do we know which teams will benefit most?

Use the scoring table above. Our free three-minute AI readiness assessment can also help.

Sources

All sources last checked 15 November 2026.

Where FuKazee fits in

FuKazee helps Kenyan organisations train the right people in the right order, starting with leaders and managers, on the work already on their desks.

Tell us your teams and headcount, and we will suggest an order that fits your operations. Email lead@fukazee.com

Want guided support instead?

Plan AI training around your team's real work. Answer a few quick questions about your sector, team size and goals, and we will recommend the right programme and format. Sessions are built on your workflows and include a usage policy.

Rodney MainaFounder & Director, FuKazee

Rodney has run FuKazee's training since 2022 and still teaches inside every programme line: corporate sessions, the postgraduate cohort and the KCSE examiner webinars. First Class Honours in Journalism and Mass Communication, University of Nairobi.