The licence rollout
Seats for Copilot or ChatGPT, a launch email, a spike in logins. By week six the usage dashboard tells the real story.
The tool works. Nobody connected it to the work.
Corporate AI training
Every employee. Every report. 3× faster.
Your people already have the AI licence. What they don't have is a method for the documents they produce every week. This is hands-on AI training for corporate teams, delivered as live workshops with packs to follow along. We time how long your people take today, train them on their own reports, proposals and board packs, then measure it again. Priced per employee, against the capacity it returns.
- Live workshops, in the room or online
- A playbook for every role
- Measured before and after
Organisations we have worked with
- Ncedo Afrika
What you've already tried
You bought the licences. The work didn't change.
Most organisations have already spent on AI. The spend shows up in the budget. It rarely shows up in how long the monthly report takes.
The lunch-and-learn
An inspiring hour on what AI can do. Everyone nods. Monday's board pack is built exactly the way it was last month.
Nobody connected the tool to the board pack.
The e-learning module
Forty minutes of generic video and a green tick in the LMS. Completion was measured. Capability wasn't.
The tool was taught. The documents were left alone.
The innovation pilot
One clever proof of concept in one team. It impressed ExCo, then never left the sandbox.
It connected the tool to a demo, not to the reports people produce every Monday.
The policy memo
AI restricted pending review. Meanwhile people paste company documents into personal accounts, because the deadline didn't wait.
The tool was locked. The risk walked right past it.
And the numbers say you're not alone.
- 95%
- of enterprise generative-AI pilots delivered no measurable impact on profit and loss.MIT NANDA, The GenAI Divide, 2025
- 39%
- of people using AI at work had received any AI training from their company.Microsoft & LinkedIn Work Trend Index, 2024
- 78%
- of AI users were bringing their own AI tools to work.Microsoft & LinkedIn Work Trend Index, 2024
- 36%
- of employees feel they have received adequate AI upskilling.BCG, AI at Work, 2026
Global research. Your organisation's own numbers come from the baseline.
The tool was never the problem.
Nobody connected it to the documents your people actually produce. Connecting it is the whole job.
The difference
Built on your documents, not the tool's feature list.
Adoption stalls when training teaches the tool and leaves the work alone. We start from the other end.
The tool already works. What doesn't work is the gap between the tool and the document your person produces every Monday morning. A monthly report doesn't need someone who knows more AI features. It needs a method: a structured brief instead of a blank page, the right source material already summarised, the house style applied in seconds instead of hours. Close that gap and the tool does what the licence promised.
That gap is where we work.
Here's what that looks like, line by line.
Typical AI trainingGrowthCred
- Starts withTypical AI training: The tool and its featuresGrowthCred: The documents your teams produce every week
- ExamplesTypical AI training: Generic prompts from another industryGrowthCred: Your own templates, reports and proposals, redacted where needed
- FormatTypical AI training: A talk, then back to workGrowthCred: Workshops where every person builds on a real document
- People leave withTypical AI training: Slides and good intentionsGrowthCred: A playbook, a prompt library and a context document for their role
- Measured byTypical AI training: Attendance and completionGrowthCred: Minutes per document, before and after
- Priced onTypical AI training: A day rateGrowthCred: Per employee, set against the capacity returned
Who runs the room
Trained by someone who builds with it every day.
Phila Ngwenya founded GrowthCred to put AI to work on the documents that run a business, not to demo it. Not to talk about what AI might do one day, but to make this month's report faster this month.
- 7
- AI talks at WeWork Johannesburg
- 5 / 5
- Seats sold at the first Operator Intensive, WeWork Rosebank
- 3 days → 2 hrs
- A TaiAscend document process, after working with us
Read the client stories“My Claude has been operating a lot better since the last session.”
The transformation
The target is 3×.
A report that takes six hours takes two. Not by learning more features, and not by skipping review, but by connecting the tool to the actual document and removing what surrounds it: the blank page, the hunt for last month's numbers, the reformatting, the third rewrite for a different audience.
One monthly report, in hours
One monthly report, in hours: 6 hours before, 2 hours after.
Illustrative. Your baseline sets the real starting point.
Where the hours come back from
- 01First drafts written from a structured brief, not a blank page
- 02Source material summarised before anyone reads all of it
- 03One document reshaped for the board, the client and the team
- 04House style and formatting applied in seconds
- 05Review checklists that catch errors before a manager does
TaiAscendClient result
3 days2 hrs
A document process that took three days now takes two hours.
Read the client storiesThe business case
What 3× is worth on your headcount.
Every month the gap stays open, your team produces the same documents at the same speed, at full salary. The licence you already paid for sits open on their screens, and the reports take as long as they always did. The cost isn't what training costs. It's what the gap costs, every month you leave it.
Move the numbers to match one team. The model is deliberately simple, so you can check it before you take it to ExCo.
One department, or the whole organisation.
Reports, proposals, board packs, client letters, minutes, SOPs.
Salary plus benefits, before tax.
Capacity returned, per year
R3.2M
- Hours returned, per year
- 12,267
- Full-time roles of capacity
- 6.7
- Value per employee, per year
- R64,000
That last figure is the ceiling on what training should cost per head. A per-employee fee well under it is the case for training; one near it is the case against.
Capacity is not cash. Saved hours only become savings if output grows or hiring slows, so we agree with you where the hours go before the first session.
Estimate from your inputs: 46 working weeks, 40-hour weeks. Not a result measured at your organisation.
For the people who sign it off
One programme. Four sets of questions answered.
CFO / FD
“What's the return, in rand?”
A rand figure, before and after.
The baseline and the thirty-day measurement are in rand per team, so the return is a number, not a feeling.
HR / L&D
“What happens when people leave?”
Capability that stays when people move.
Every role has a written playbook, so new hires inherit the method on day one instead of waiting for the next training budget.
COO / Heads of department
“Which documents get faster?”
Shorter cycles on the documents that hold up decisions.
The reports, packs and proposals your decisions wait for are the ones we train on first. You name them at the baseline.
IT / Risk
“What about our data?”
Inside the tools you already approved.
We train on your sanctioned platform, practise on redacted material, and write what may never go into an AI tool into every pack.
Delivery
Workshops your people build in. Packs they keep.
- 01
Baseline
We time a sample of the documents each team produces today: monthly reports, proposals, board packs, client letters. That is the number we are measured against.
- 02
Live workshops
Presentation-style sessions, in the room or online, run team by team. Every person works on a real document from their own desk, not a demo on ours.
- 03
Follow-along packs
Each role gets a printed and digital pack to follow during the session and keep afterwards, so the method outlives the day.
- 04
Measure and report
Thirty days later we time the same documents again and report to the sponsor: hours returned, by team, in rand.
What every employee leaves with
Nobody goes back to a blank page.
- A step-by-step playbookThe method for their role, written for the AI tool your organisation has approved.
- A prompt libraryReady briefs for the ten documents their role produces most.
- A team context documentHow your organisation works, who it serves and how it writes, so every draft starts informed.
- A review checklistWhat must be checked before anything leaves the building, and what never goes into an AI tool.
- Their own before and afterThe time they took on a real document at the baseline, and the time they take now.
Pricing
Priced per employee. Anchored to what it returns.
You already paid for the tool. This is the missing piece: the method that connects it to the work. You pay per person trained, so the investment scales with the headcount that produces the return. We quote after the baseline, so the per-employee fee sits beside the per-employee value in the same proposal. If the numbers don't make the case, you'll see that before you sign.
Per employee01
Pilot team
One department, one document type
Prove the number on one team before anyone commits to a rollout.
Request a proposal for a pilot teamPer employee02
Department rollout
Several teams, a pack per role
The documents each team produces most, trained team by team, measured team by team.
Request a proposal for a department rolloutPer employee03
Organisation-wide
Multiple departments or sites
Phased across the business, with the baseline and measurement reported to one sponsor.
Request a proposal for a organisation-wideEvery tier is priced per employee. The rate is set in the proposal, after the baseline.
Every week this stays unfixed, your people open the AI tool, stare at it, and go back to the way they've always done it. The licence renews. The reports stay slow.
And odds are that right now someone on your team is pasting a client document into a personal AI account, because the deadline didn't wait for the training budget.
The tool was never the problem. The connection was. This is the connection.
Next step
Request a team proposal.
Tell us the team and the documents that slow it down. We'll come back with a baseline plan and a per-employee quote.
Tell us about the team
One confirmation email, one reply. No list, no newsletter.
Questions
Questions sponsors ask first.
What does corporate AI training from GrowthCred include?
A baseline of how long your teams take on their real documents, live workshop sessions run team by team, a follow-along pack for every role, and a measurement thirty days later reported to your sponsor in hours and rand.
How do you measure 3× faster?
Before training we time a sample of real documents per team, at the review standard you already use. Thirty days after, we time the same kinds of document again. 3× is the target we train to; the report shows the measured number, whatever it is.
How is corporate AI training priced?
Per employee trained. We quote after the baseline, so the fee is set against the capacity it returns. The number of people, teams and role-specific packs changes the quote.
Which AI tools do you train on?
Whichever platform your organisation has approved, such as Microsoft Copilot, ChatGPT, Claude or Gemini. The method stays the same; the packs are written for your tool.
Is the training in person or online?
Either. Sessions run in the room in Johannesburg, or online for teams elsewhere in South Africa and abroad.
What about confidential information and POPIA?
Practice uses redacted or fictional material, and every pack sets out what may never go into an AI tool. Training happens inside your approved, company-controlled accounts. Training alone does not make a workflow compliant; your policies and legal team still decide.
Do our people need to be technical?
No. If they can write an email and work in a spreadsheet, they can do this. Every session is built around documents they already produce.
Is the training accredited?
No. This is practical workplace training, not an accredited qualification, and no accreditation is claimed.