You don ever think about wetin dey different between tool and system?
Tools na wetin we dey use when we dey work. Systems too, dem dey make work dey easy.
Some people fit think say system na just tool wey big pass.
But if we divide work into two kinds—work wey you dey do small small and work wey dey follow step-by-step—the difference between tool and system go clear well well.
Iteration And Flow
Iterative work na the way you dey do something small small, dey try and fail, dey change am as you dey go, until you finish am.
For this kind work, one bag full of different tools wey you fit pick from based on wetin you wan do, go dey very helpful.
Flow-based work, for the other hand, na when you dey follow steps one by one, until you reach the last step to bring out wetin you dey do.
For flow-based work, if you get a system wey dey guide you through the steps, your work go fast well well and the quality go better.
Changing Iterative Work To Flow-Based Work Plus Systematization
Plenty work wey human beings dey do, na either iterative work, or e be part of one flow-based work wey don get system already.
If you change iterative work to flow-based work, and then set am up as a system, e go really help to make work fast and better.
The Industrial Revolution And The IT Revolution
The Industrial Revolution and the IT Revolution na clear examples of how work became faster and better quality, all because iterative work was changed into flow-based work and then made into a system.
Before the Industrial Revolution, people dey produce things small-small, like iterative work. Dem dey use tools well-well, and dem fit change how dem arrange things or how dem do am anytime dem want.
Before the IT Revolution, how dem dey process information too na iterative work. Humans dey use tools and dey do am anyhow dem like, no fixed way.
But when dem make all these processes into a system, just like factory production lines and business IT systems, work became faster and better.
However, no be only to make am a system, but to change that iterative work into flow-based work na very important. Na because dem first change am to flow-based work, systemization come possible.
The Generative AI Revolution
When person wan use generative AI for business to make work fast and good, just to use AI like say na normal tool no go give real value.
The main thing wey we dey target na to change iterative work into flow-based work, then make that flow-based work into a system.
Generative AI, because e fit adapt quick-quick, fit handle tasks wey be iterative. But whether na human or generative AI dey do am, iterative work get limit to how fast and how good e fit be.
Na why e important to aim for flow-based transformation and systemization.
Person fit argue say if flow-based transformation fit make work fast and good even with human workers, then dem for don do am before generative AI come out.
But to change work to flow-based based on say human dey do am, na actually very hard problem. Human workers no fit quick quick adjust to changes for wetin dem suppose do or how dem suppose do am.
On the other hand, if na generative AI be the worker, e easy to change wetin e suppose do and how e suppose do am by trying different ways.
Unlike human beings, generative AI fit forget wetin e do before, read and understand new steps for one second, and work based on dem.
So, the main way to use generative AI for business go be to change iterative work to flow-based work and then make am into a system.
How To Make Business Dey Fast Using Generative AI
Make we look example of how generative AI fit make business dey fast.
For example, think about how dem dey answer staff questions about company rules.
If you use generative AI, e fit search company rules and write out possible answers.
But, e fit be say the generative AI go use old rules or just sabi guess and give answers wey no dey inside the rules clearly.
Again, questions dey come for different ways, like email, chat apps, phone calls, or even just by mouth.
So, staff wey dey handle questions still need to collect dem like before.
E possible say work go dey fast if dem answer questions wey dem fit answer sharp-sharp, and for questions wey need rule checking, dem fit put the question inside generative AI to get a draft answer.
Plus, for questions wey dem dey ask plenty, e good to put dem for the company internal website as FAQs (Frequently Asked Questions).
Generative AI fit also help by taking normal questions and answers, then turn dem into bullet points wey fit dey published on website.
Apart from that, when dem need to change rules, generative AI fit help to draft proposals.
These kind uses fit make some percentage of the question handling work faster.
But, this one just mean say the question handling still be iterative work, and dem just dey use generative AI like a tool.
Because of that, the level of efficiency increase go just be small.
Flow-Based Work Transformation
To make the inquiry handling task, wey we use as example, work for the best way, dem must change am to a flow.
To do this one, the work wey the person in charge dey do when e dey handle inquiries need to dey detailed and put for formal way.
- Collect questions through different different ways.
- If the question be like the one dem don answer before, and the rules wey connect to am no change, give the same answer.
- For new questions or questions wey rule don change, check the rules and write out a draft answer.
- Check say the draft answer no talk about old rules or put something wey no dey inside the rules.
- Check if you need approval before you answer, and get the approval if e necessary.
- Answer through the same way dem send the question.
- Record the question matter, the approval result, and the answer result inside the inquiry history data.
- Dey check the inquiry history data regularly and create drafts to update questions and answers wey dem dey ask often.
- Update the company's internal homepage after you get approval.
- Update the rule data wey dem dey refer to when rules change.
- At the same time, record inside the past inquiry history data say related answers and rule updates don happen.
- Confirm if frequently asked questions and answers need to dey reviewed because rule don change, and update am if e necessary.
When you define the details of the work wey dem dey do clearly, like we talk above, you fit connect these tasks, and change flexible iterative work into a clearer flow-based process.
Example Of Systematization
When you create this kind work-flow, the road to make am a system go clear.
For systematization, if e no bad to manage small inconvenience for staff, one way na to make all the channels wey questions dey enter become one.
But if staff comfort na the main thing, then the system suppose fit collect questions from all channels.
Basically, the system suppose collect questions directly. Na only for questions wey dem ask by mouth, the person in charge go put am inside the system.
After the system collect a question, the IT system and generative AI suppose do plenty of the next steps as much as possible, following the flow. At first, human checks and approvals suppose dey scattered inside the system, and human operators suppose fit correct things.
Then, as dem dey use the system to handle questions, if generative AI make mistake, dem suppose update the instructions for the generative AI with warnings, things to check, examples of mistakes, and correct examples to stop am from happening again.
This go help to reduce mistakes from generative AI. This process of updating instructions for generative AI fit even dey more efficient if dem change am to a flow-based task instead of an iterative one.
Na so, by making flow-based work into a system, even work wey look like say human must do am fit dey replaced by a system wey generative AI dey control.
Wetin People Dey Always Misunderstand
Plenty people dey think say to use generative AI for business no too work well now, or say e still too early.
But, many of dem dey fall into two kind of misunderstanding.
The first misunderstanding na because dem dey focus on just using generative AI like a tool.
As we don show here, to use generative AI like a tool for iterative tasks no dey really make business dey fast well well. When dem see am or hear am, e dey cause this misunderstanding.
The second misunderstanding na because dem dey focus on making generative AI do iterative tasks.
True true, if you try make the generative AI wey dey now do iterative tasks, e no dey work well. So, generative AI no fit totally do all the work wey humans dey do, and if you just focus on this point, e go lead to misunderstanding.
At Last
As we don talk so, if you change iterative work to flow-based work and then make am a system, you go see better results pass just using tools.
Plus, even if generative AI no fit do all iterative work by itself, e fit handle many individual tasks inside a flow-based process now. Even if e make plenty mistakes at first, you fit dey improve am small small by updating the instructions you dey give am.
Another option na to divide tasks as e necessary. You fit separate the drafting work from the checking work, or even do checking for many stages.
If you fit make things a system like this, then improvements go dey happen with every task, and work go dey easier and faster as time dey go.
This na one way to work wey go make the system itself dey improve continuously, just like how factory production and IT system implementation dey be.
To truly use generative AI well, you gats change how you dey think: instead of just making your own iterative tasks better, you need to look at your work well well and change am into flow-based processes, then make dem into systems.