My client (a French bank on-line) would like to analyze keywords used in the Lithium community.
They would like to identify topics with keywords (for instance: CB, payment, purchase, road toll…) and know the number of replies and the number of members who participate to this discussion. With this approach, my client would like to improve process treatment of negative posts and also, address directly community members who complaint.
Unfortunately, our system is not offering this feature. Among our clients we have specialists in this topic such as SAS Institute, Alteryx, RapidMiner, Microsoft.
I was wondering if you could have some client’s examples of semantic tool integration? And perhaps, some best practices to share.
We currently use a social listening tool called Brandwatch which does this quite well. In addition to covering public content on our community it also scans the wider social space to build a much more complete picture.
On the Lithium side, we have our community integrated into Lithium Social Media Management, this allows us to tag all discussions on the community based on language and keywords similar to the social media workflow. Using these tags we're able to build work queues if needed with prioritisation and smart views in Lithium SMM Analytics to monitor volumes.
Lithium Social Intelligence does provide some context and insight as well into the community, it's really useful for the analytics around threads/discussions. There's a mixture of options available to tackle this one.
I'm sure other members of the community will have examples to share in this space as well. Happy to discuss further if you want to go into a bit more detail on what our approach is.
The Community that I used to manage has an integration with Clarabridge to analyze data and sentiment. It was useful for the organization because it was integrated within many aspects of the product, so putting it to use in the Community made sense - we were able to compare Community sentiment and keyword tracking throughout the multiple touch points we had with a customer.
The piece of advice I would give is that you are only as good as the data you get - if you are only looking at one touch point (Community) then that is all you are going to be able to analyze. Don't get me wrong, Community is a great place to start, but having a more robust view is a great end goal and could help get more data to analyze different touch points. With more touch points and tracking you can start to find trends and potentially get to a point to be more proactive within your Community to try to deflect/deter some negative sentiment before it ever happens (for example around releases, updates, policy changes, etc.).