Your Organization Doesn’t Need a Chatbot. It Needs Its Hours Back.

It’s Monday morning in a synagogue office somewhere in Los Angeles. The administrator exports last week’s event registrations from one system. She pastes them into a spreadsheet, fixes the names that came through in all caps and retypes the new families into the membership database. Then she emails the caterer a headcount and updates the board on how many seats are left.

By lunch she has done three hours of work that a computer should have done in three seconds.

Every community organization in this city has a version of that Monday. So do most family businesses, schools and agencies. And while the headlines about artificial intelligence focus on chatbots and robots, this is where the real opportunity sits: in the quiet, repetitive work that eats the hours of the people you already have.

What changed

Most of the software organizations that already use (email platforms, donor databases, registration forms, accounting systems) has always been able to talk to other software. The technical term is an API: a door that lets one system send information to another automatically.

The problem was that walking through that door required a developer. Connecting two systems meant a scoping document, an invoice and a wait measured in months. So most small organizations never did it, and hired people to be the connection instead.

That has changed faster than most leaders realize. Automation tools like Zapier and Make can link hundreds of common systems without code. And AI coding assistants now let one person who understands the tools build in days what once took a development team months. The cost of connecting your systems has collapsed. Most organizations just haven’t noticed yet.

Where to start: the task someone dreads

Don’t start with technology. Start with a question for every person on your staff: what do you do every week that you wish a machine would do for you?

The answers are almost always the same kind of thing. Exporting a list. Retyping names. Reconciling two spreadsheets that should agree and don’t. Chasing the same five people for the same form. Each of those is a candidate.

Then follow a single name through your organization. A family fills out a form for the Purim carnival. Where does their information go next? Into an email list? A database? A spreadsheet for the volunteers? Every place where a human copies that name from one screen to another is a place where an automation belongs, and a place where a typo is waiting to happen.

Pick one. The most annoying, most repetitive, lowest-risk task on the list. Fix that first, and let the staff member who used to do it feel the difference. Nothing builds support for the next project like getting a Monday morning back.

What it looks like in practice

None of this requires a futuristic system. Here are the kinds of changes that make the biggest difference for small teams.

The weekly list that builds itself. Many organizations have someone who spends half a day each week exporting, cleaning and sending a list to a vendor, a printer or a volunteer team. That job can run itself overnight, checked by the same rules the staff member used to apply by hand.

Registrations that go where they belong. When someone signs up for an open house, a class or a High Holiday service, their details can flow straight into your database and email list, tagged correctly, with a confirmation sent. No retyping, no lost forms.

First drafts in minutes. AI tools are genuinely good at first drafts: a board summary, a thank-you note, a volunteer update. In a city where many community members read Hebrew, Farsi, Russian or Spanish more comfortably than English, AI can also produce a solid first translation of a newsletter. A person should always review it, but the blank page is gone.

Numbers that update themselves. Instead of a monthly scramble to assemble enrollment, membership or attendance figures, a simple dashboard can pull them from the systems you already use and stay current every day.

What not to do

Don’t start by buying software. A new platform won’t fix a process nobody has written down. Map the work first; the right tools usually become obvious, and you often already own them.

Don’t build something only one person understands. Every organization has a system that “only the volunteer’s nephew knows how to fix.” Document every automation in plain language, and make sure at least two people know it exists and what to do when it breaks.

Don’t take humans out of the sensitive parts. Anything that touches money, donors, members’ personal lives or pastoral care deserves a person’s eyes. Automate the copying, not the judgment.

Don’t paste private information into free AI tools. Member lists, donor records and personal details don’t belong in a consumer chatbot. Use tools with proper business privacy terms, and set a clear rule for staff before someone learns it the hard way.

You don’t have to finish the work

We are a community that already understands the value of setting work down. The point of all this isn’t to replace the people who keep our institutions running. It’s to give them back the hours they spend being human copy machines, so they can spend those hours on the work only people can do: welcoming a new family, calling a donor, noticing who hasn’t come back.

Pirkei Avot teaches that it is not your duty to finish the work, but neither are you free to desist from it. That’s the right way to approach AI in a small organization. You don’t need a grand strategy or a big budget. You need one task, one week and the willingness to start. 

Lior Ron is the founder and principal of Lightdrop, a Los Angeles marketing and growth agency founded in 2011. Lightdrop builds automations, AI tools and digital strategy for businesses and nonprofits, and leads the Jewish Journal’s digital operations.