---
title: "Launch is not the end."
description: "With monitoring, hosting and support, we keep your chatbot or voicebot running, even as more people use it."
url: https://www.lautmaler.de/en/services/operations
lang: "en"
alternate_de: "https://www.lautmaler.de/services/betrieb"
---
[Services](https://www.lautmaler.de/en/services) Operations

05 / Operations

# Launch is not the end.

With monitoring, hosting and support, we keep your chatbot or voicebot running, even as more people use it.

[Discuss a project](https://www.lautmaler.de/en/contact)[See all services](https://www.lautmaler.de/en/services)

![Illustration: Monitoring-Dashboard mit Diagrammen und Kennzahlen, davor Telefon und Tablet](https://www.lautmaler.de/uploads/MoCcEcZ58Jt_L56NvrHwg-svc-operations-scene.png)

Case study

## In practice

![sipgate-Logo vor einem Büro, in dem Mitarbeitende an Schreibtischen arbeiten](https://www.lautmaler.de/uploads/x0B2c1oyAyvarP4tOz6ee-sipgate-hero.jpg)

[Discuss a project](https://www.lautmaler.de/en/contact)[View the sipgate project](https://www.lautmaler.de/en/projects/sipgate)

sipgate

/ Customer service automation

Today more contacts go through the bot than through the hotline.

We have been running sipgate's chatbots since 2020. A simple Dialogflow bot grew into Botify, the in-house platform behind all sipgate bots. We continuously analyse real, anonymised conversations, retrain and roll out improvements in a controlled way. Today more than a third of all contacts go through the bot, over 3,800 a month.

9%

of their profit is what the world's largest companies lose to **unplanned outages**.

38%

can **always determine the cause of an outage**. Others find it only with difficulty, or never.

90%

see a **rush on support** after an outage.

/ What's included

## What our chat- and voicebot monitoring includes

We take the bot live, host it, monitor it and report to you on how it's doing. When usage grows, the infrastructure grows with it.

### Integration & hosting

We integrate the bot into your existing IT. We set up chatbot hosting to your data protection and compliance requirements, in the cloud or on-premise on your own servers. Security and performance both have to be right.

### Deployment & go-live

We're there at go-live so the switch from testing to live operation goes smoothly. For models and NLU we build CI/CD pipelines, automated processes through which updates go live quickly and safely.

### Monitoring & KPIs

We set up a chatbot dashboard that shows how your bot is doing at any time. You see problems there before anything fails. With LLM monitoring we also check that the answers are correct. We agree the KPIs you measure success by together.

### Operations, support & scaling

After launch we stay responsible and take care of technical support. When usage grows, we scale the infrastructure before it hits its limits.

### Stakeholder reporting

We report regularly on status and metrics so all stakeholders keep the overview.

/ FAQ

## Frequently Asked Questions

Here you'll find the questions we're asked most often. Can't find what you're looking for? **Then just chat with our bot!**

[ChatLord](https://www.lautmaler.de/en/contact)

**How do you measure the application's success?**

By business KPIs that we set together with you. One of the most important is the containment rate, the share of requests the bot resolves entirely without a human. Whether the bot helps users along the way is something we check in parallel in research.

**Can we choose between cloud and on-premise hosting?**

Yes. We set up chatbot hosting to your data protection and compliance requirements, in the cloud or on-premise on your servers. Security and performance have to be right either way.

**How do we know whether the application is running stably?**

Through a monitoring dashboard that shows the bot's status at any time. Problems show up there before anything fails.

**What happens if our user numbers grow sharply?**

Then we scale the infrastructure with them before it reaches its limits.

**Does your responsibility end at launch?**

No. If you want, we stay on board with technical support when questions or problems come up after go-live. We've been doing this from Berlin since 2007, for companies such as sipgate and Telekom.

**What does it cost to run a chatbot per month?**

It depends :). Mainly on four things: the hosting, how many conversations the bot has, the LLM costs per conversation and how much support you need. We work out the figures with you before launch, so there are no surprises once it's running.

## Going live is easy. Staying live is the real art.

![Co-Founder & Managing Director Technology - Implementation, Information Architecture, Backend Integration, API Integration, Data Visualisation, Business Consulting](https://www.lautmaler.de/uploads/2Knki6nQUDF6XrijJFQL3-team-core-1.webp)

Your contact

Norbert Gocht

/ Co-Founder & Managing Director Technology

[Contact](https://www.lautmaler.de/en/contact)
