8 things we have learned building apps for client-heavy businesses

We build apps mostly for businesses that deal with the same clients for months or years. The client we can say most about in public is Merezha Prava, a Ukrainian legal-services company. We built and ran its client app for about two years, and the first MVP shipped in 3–4 months.

Our founder wrote the eight lessons below on LinkedIn after dozens of calls with founders. Here each lesson comes with one fact from a project we can name.

1. Repeat contact matters more than your industry

An app doesn't make sense because of your sector. It makes sense when the same client keeps coming back with a question, a document, a payment or a status check.

From our work: two of our cases have nothing in common on paper. One is a Ukrainian legal-services firm. The other is Dutch Bros, a US drive-thru coffee chain, where one of our Android developers worked inside the client's team on Dutch Rewards loyalty features. The product and its results are the client's; our part was the engineering role. See the Dutch Bros case.

2. Below about 500 cases a month, an app rarely pays for itself

This is our founder's rule of thumb from scoping calls, not a formula. Under that volume, a phone and a spreadsheet usually cost less than building and running an app.

From our work: Merezha Prava is far above that line. Its app has 36,000+ users.

Don't take our rule of thumb on trust. Put your monthly inquiries and call volume into the ROI calculator. It tells you when your volume is too low for a full build. We have turned down potential clients many times because their business did not need an app; a small restaurant chain was one of them.

3. If the only problem is answering questions, you need a chatbot, not an app

We say this even though apps are our business. As our founder put it in a later post, a bot closes the FAQ layer. But it can't answer "what's happening with my case?" without access to the case system. An app earns its cost when clients need to do something.

From our work: the Merezha Prava app is built around those actions. Clients follow case stages, payments, documents and lawyer comments in one place, instead of calling the office for each update. See the Merezha Prava case.

4. The app won't earn money directly. It stops you paying for manual work

Count it as a cost you remove, not as new revenue. The cost is staff time spent on status calls, chasers and documents asked for twice.

From our work: at Merezha Prava, automatic case-stage notifications cut inbound status-check calls and the information department's workload. The firm gets 15% fewer inbound calls.

The ROI calculator prices each simple-question call an app answers at £5.58. That is ContactBabel's published UK cost of handling one inbound call (2024 guide). We walk through the full maths in App ROI is the labor you stop paying for.

5. Building it is the easy part. Getting clients to open it is harder

The feature our founder underestimated most was document upload. He never expected it to take off the way it did. It replaces a clumsy step: sending passports and contracts through chat apps, where the file is gone once a client changes phones.

From our work: Merezha Prava's clients took up in-app document exchange in the first months after release. The case page counts 36,700+ documents uploaded in the first six months.

6. A website waits to be remembered. An app sends a notification

A client has to decide to check a website. A push notification reaches them before they pick up the phone to ask.

From our work: legal clients are supposed to be privacy-cautious. Yet 77% of Merezha Prava's app users turned push on. The app notifies them at each stage of the lawyer's work.

7. Some cases are too human to automate. Leave them alone

Not every conversation belongs in the app. Part of honest scoping is naming the ones that should stay a phone call or a meeting.

[TEAM SUPPLIES: one interaction a named client deliberately kept out of the app, and why; needs the client's OK. Candidate: which Merezha Prava steps still happen by phone or in the office]

8. Without an app, more clients means more staff. With one, it doesn't have to

An app lets a client-heavy team take on more clients without hiring at the same rate. The return shows up as hires you don't have to make.

From our work: this was the Merezha Prava brief. Thousands of clients needed clear case status without overloading the information department. [TEAM SUPPLIES: how the information department's size changed as the client base grew; only with Merezha Prava's OK]

When it's worth a conversation

We talk founders out of apps more often than into them. If your process repeats and your team is stretched, that is the time to talk. Not before.

Before you call, the app cost calculator shows what a first version would cost. The ROI calculator shows whether the hours it saves would pay that back. If both look right, see how we run new app and MVP projects, or book an app ROI review.

The original post

As Oleksiy Kazavchynskyy posted it on LinkedIn on 1 July 2026, unchanged:

8 things I've learned building apps for client-heavy businesses:

I run a dev shop that builds apps for legal and client-heavy firms. Lately I've spent dozens of calls with their founders. Here's what keeps coming up.

1. The industry doesn't matter. What matters is whether the same client comes back to you again and again.
2. Below 500 cases a month, an app won't pay for itself. A phone and a spreadsheet will do the job cheaper.
3. If the only problem is answering questions, it's a chatbot, not an app.
4. The app won't make you money directly. It saves the money you're already losing on manual work.
5. The hard part isn't building it. It's getting clients to open it. The feature I underestimated most was document upload. I never expected it to take off the way it did.
6. Clients have to remember to check a website. An app notifies them. They find out before they pick up the phone to ask.
7. Some cases are too human to automate. Know which ones, and leave them alone.
8. Without an app, more clients just means hiring more people. The app lets you handle more clients without growing the team.

I talk founders out of apps more often than into them.

If your process repeats and your team is drowning, that's when it's worth a conversation.