This update is another step towards bringing our users more transparency on how Lang works. In a nutshell, hovering on a tag in the Search view will highlight the piece of text that is relevant for the tag.

Over the past month we have been working on enhancing Lang's experience to make it easier to use while being able to get granular insights. We're excited to announce a major update coming in November!

The alert creation feature will send you notifications when a selected tag follows a predefined behavior, allowing you to react more quickly to critical issues as well as to identify relevant changes that can lead to actionable insights.

Our first version allows to create alerts in the Setup section and receive them by email notifications. The current alert configuration consists in selecting the tag to configure and setting the condition which will trigger the alert, based on a volume of documents with the selected tag.

Note: current tag alerts configuration follows the logic "notify me when there is an increase of X (absolute value) number of documents / tickets". Every hour, the model calculates the amount of tickets processed during the last hour, and if the amount matches the condition, an alert is sent.

Once the initial implementation stage is complete, one of the main concerns is how to efficiently maintain the project, keeping a high level of coverage and identifying opportunities to improve tagging and automation.

Lang Optimize proactively flags the workflows you need to review and the tags you need to build to reduce operational bottlenecks and focus your time less on manual tasks and more on your customers.

This functionality performs periodic tag quality assessments to discover potential errors or sub-optimal tag configurations. Currently it offers two different types of audits.

On the one hand, there are the recommendations for incorrect tagging. The goal of this audit is to show the number of tickets that had a certain tag and no longer have it because an agent manually removed it.

On the other hand are the overlapping recommendations. In this case the goal is to show the overlapping of tags to detect potential problems when defining the tags, or the need to update a tag to be more specific.

In this section you will find the top 5 recommendations for each of these audits. You will also be able to see a sample of the documents that are affected to more quickly understand where the problem is and what action to take.

This feature allows you to discover new and trending concepts that are emerging from new tickets / documents that are incorporated into the project.

The focus when generating new concepts is placed on tickets that do not have tags. The goal is to allow you to tag untagged documents and thus increase coverage.

To enable this functionality you will have to reach out the Lang team. Once enabled, you can change the settings of this functionality from the Setup > Optimize section.

This functionality analyzes your data and looks for patterns in tickets that have been automated in order to offer you recommendations on tags to apply automations to. Currently there are routing and reply recommendations.

You can change the settings of this functionality from the Setup > Optimize section. This will allow us to generate recommendations that are better suited to your needs and that better reflect the reality of your business.

We know you have a lot of tasks to pay attention to, so at Lang we want to give you a faster and more proactive way to learn about new recommendations we generate. By subscribing to Optimize notifications, we will inform you via email when new recommendations are generated. In this way you will not have to worry about missing recommendations; we will notify you so that you can review them as soon as there is news.

To configure notifications, you have to go to the Setup > Email notifications section and enter the emails of the people you want to receive the alerts in the "Optimize" block.

From now on we will only send a single automatic reply when there is more than one tag with reply automation applied to a ticket. In this way, we prevent your customers from receiving multiple automatic responses to the same ticket, which often causes confusion or unsatisfactory experiences.

Take into account the following:

One of the most recent features we released is multi-projects. This is a new type of project that allows you to aggregate other projects to see trends, make comparisons and get other insights in a place that has capabilities very similar to those of the Dashboard section in any traditional project.

Multi-projects can be very useful if you have different projects by country, by brand, by product, etc. and you want to see the aggregated data in one place.

As mentioned above, multi-projects displays data in a section that works similar to the Dashboard section in any other project. The difference is that at the moment in a multi-project you will not be able to:

In a multi-project you can filter by metadata. To choose which metadata to use, you have to go to the Setup section. There you will see all the metadata of the documents belonging to the child projects.

Finally, here are other things to keep in mind:

We redesigned the projects table so that the dropdown that appears under global projects shows information related to child projects. In this way, it is easier to understand the relationship between the projects.

The project details that previously appeared in that dropdown can now be seen by clicking on the "See project details" option in the Actions column.

In the Automations section, you can now see the average reply time and the average first resolution time of the tickets associated with a tag. We believe these metrics can help you understand the impact of your actions in Lang, as well as to identify opportunities for improvement.