Lily’s Beginnings: Building the Care Behind the Service

Every visit to a home begins long before anyone arrives. Someone has agreed on what the service includes. Someone has understood what the property needs. Someone knows who is responsible, and how everyone will know the work is finished well.

Those quiet steps are what I have been thinking about most while building Lily. We are developing our operating processes at the same time as our website and client portal, and I want each of them to reflect the care we hope clients feel in their homes.

This is an early account of that thinking. It describes how we are approaching the work, not results we have already proven.

Understanding the work first

Before asking what a tool should do, I want to understand what each service asks of the people who deliver it. That means the physical work, and also the information and decisions around it.

A service description has to become something a person can carry out. A client’s preference has to reach the person who will act on it. A change has to be understood before anyone follows an outdated instruction. When the work behind a service is unclear, a more convenient way to request it does not make delivering it any clearer.

As of October 17, 2026, the point this account describes, onboarding at Lily was handled manually, the client portal was launching, and no clients had used it yet. My priorities were understanding labor, improving booking, and making sound decisions about cost.

Why small tasks matter

Some of the most important moments in property care are small. As an illustration, consider confirming exactly what a client would like included in a visit. It may be a short exchange, yet the answer shapes who is assigned, what they are asked to do, what supplies are needed, and what the client expects to see at the end.

The same is true of carrying a property’s particular requirements into clear instructions, or recording a correction a client has asked for so it is not lost before the next visit. These are examples of the design questions I am working through, not descriptions of finished features.

For each of these moments, I want to know what starts it, who is responsible for it, and what becomes possible once it is done. A small task is worth attention when important work depends on it.

Clarity before software

Making a process clear is a design question in its own right, and it comes before deciding whether software should support it. Sometimes the better answer is a clearer service description, a consistent way of writing instructions, or one fewer question during onboarding.

Every feature also asks something of people. Someone has to enter the information, keep it current, and know what to do when it is wrong. I want the reason for building something to be strong enough to justify that ongoing work.

Booking is a good example. A request and a commitment are not the same thing. Ways for our team to schedule and accept work were still under consideration at the time of this account, and any design will need to say plainly what acceptance means and what happens if no one is available.

Where AI helps, and where people stay responsible

AI-assisted development is helping me build Lily’s website and portal. It can help specify and implement a tool far more quickly than I could alone. It does not decide what the business needs.

People remain responsible for service decisions, for the quality of the work, and for our relationships with clients. A screen that works still needs to be checked against how we intend to care for a home.

What finished should mean

For Lily, the measure of a visit completed well is the client’s approval. At the time of this account, how that approval would be requested and recorded had not yet been defined. That leaves a thoughtful question to answer: how do we distinguish work shared for review, work accepted, and work that needs another look? Client approval should never replace our own review of quality, and there must always be room for a conversation when something needs care.

Documenting the process

I am documenting this reasoning as the business takes shape through LiquidIQ, a research practice I also founded, in an ongoing case study called From Friction to Capacity: Building an AI-Assisted Operating System for a Small Service Business. It records the questions guiding the build, the alternatives considered, and how decisions become processes or tools.

It is a founder’s case study of my own businesses, not an independently validated assessment. Whether these choices make the work easier, the service better, or the business more effective is still to be established through use. AI assisted in developing and preparing this article.

Keeping the care human

The work behind a calm home is mostly invisible when it is done well. I would like Lily to understand that work clearly, so the tools we build serve the people doing it rather than ask more of them. Understanding the work comes first. The care itself will always belong to people.