見積もり6 min read
開発の見積もりが会社によって2〜3倍も違う理由
開発費が変わる基準と、見積書を比較するときに確認したいポイント
Read more →Twenty years of running a business, pointed at the next one.
You know your business.We know how to turn itinto something digital.
You already know where the work breaks, and you have ideas. What is hard to see is how AI and software actually turn either one into a product.
You do not need a specification. We begin by deciding, together, what is worth building at all.
No documents required · 60 minutes, online · Japanese, Korean or English
BEFORE WE TALK
Have you caught yourself thinking any of these?
"How much longer will a person have to do this?"
Phone calls, spreadsheets, email, the same manual step repeated all day.
"AI could probably handle this — probably."
You hear about AI constantly, but not what it would change in your company.
"I have the idea. I have no idea where to start."
You can see what is needed; the path from there to a service is missing.
"What am I even supposed to tell a development firm?"
There is no requirements document, and no plan to write one.
That is fine.Sorting that out is where our work starts.
WHAT YOU ALREADY HAVE
Not knowing IT is not the problem.
Twenty or thirty years in one industry gives you something a software company cannot easily acquire: experience of the actual work. The only gap is knowing how to move that experience into software.
Your side
Industry experience · Knowledge of the floor · Customers · Problems · Ideas
Our side
Framing the problem · Service design · Applying AI · Design · Build · Validation
A new digital business, and work that runs differently
Knowledge of the floor and analysis by machine, combined until it holds a business shape.
HOW WE START
We do not start building straight away.
When someone says "build me this", we do not answer with a quote. We ask three things first.
Client
"Please build this for us."
Rapid Development
- 01
"Why is it needed?"
Understanding why it is needed comes before deciding what to build.
- 02
"What is actually needed here?"
We do not build every request. We find the problem that matters most and start there.
- 03
"Let us test a small piece first."
Before large money is committed, we build the most important part for real and check it.
Building the wrong thing quickly helps no one.
Knowing what should be built comes first.
BEFORE / AFTER
Could any of this work at our company?
BEFORE
Orders arrive every day by phone, fax and email.
AFTER
Customers submit online and the orders organise themselves.
Where AI belongs — that is part of what we work out together.
PROCESS
We never open with a large build.
Four stages, and at the end of each one, stopping is a normal outcome.
STEP 01 · 60-minute conversation
We listen.
Your business, and whatever is weighing on it right now.
Decision point — stopping here is a perfectly good outcome.
STEP 02 · Diagnosis
We decide what should be done.
What to build, and just as importantly what not to build.
Decision point — the findings are yours to keep. Diagnosis alone is fine.
STEP 03 · Validation
We build something small and check it.
One critical piece, built for real and tested where the work happens.
Decision point — whether to continue is entirely your call.
STEP 04 · Build and operate
We build properly — only what survived the check.
Full development, then continuous improvement against real usage data.
FIT
Who we work best with.
GOOD ○
- Knows their industry and their business well
- Has a real problem on the floor they want solved
- Is willing to try an approach nobody has tried here
- Can take a specialist's opinion on what they do not know
- Wants to grow it while checking, not all at once
- Has the decision-maker in the room
Probably too early —
- "AI is everywhere, build us something with it."
- There is no problem anyone is trying to solve
- Every judgement and every risk is to sit with the vendor
- The goal and the direction keep moving
- Every feature must ship complete on day one
If we are not the right fit, we will say so.
Sometimes the better answer is to build nothing at all.
CASES
Less about what we built,
more about which problem it solved.
PRICING
First we decide whether
your problem should be built at all.
We will not tell you it is impossible to say until we have quoted. The price of the smallest step is published here.
Most chosen
Diagnosis and design
Decide what should be built
¥500,000
- A diagnosis of the work as it is done today
- What to build, and what not to build
- First pass at the screen flow and UI
- A rough estimate for the full build
- Take it to another firm if you prefer
Validation
Test one assumption for real
Quoted after the conversation
- Scope fixed from the diagnosis findings
- A first version that genuinely runs
- Works on mobile
- Analytics instrumented
- One month of fixes after launch
Build and operate
Grow the validated thing into a business
Quoted after the conversation
- Full development, once validation has passed
- Monthly improvements and new features
- Analytics reporting
- Unlimited consultation, answered within 24 hours
- Support for running the blog and content
※ Prices exclude tax. ※ The 60-minute conversation is free, as often as you like. ※ You can stop at the end of any stage. There is no additional invoice.
If you would rather read a little before talking
Worth reading before you commission anything
Wondering how this applies to your company?
Bringing it half-formed is completely fine.
FAQ
Questions we get often
Not at all. What we want to hear about is your business, not technology.
CONTACT
Start with the business you have been running for a decade.
You do not need to know what to build, and you do not need a document. One "this part really should change" is enough. We will start there.
60 minutes · Free · Online · Japanese, Korean or English