When available for your pricing plan, the chart builder can be found in the navigation menu on the left when you click the Insights tab.
The Chart Builder is a powerful tool for getting more insights into metrics (i.e. 'number of mentions', 'number of followers/fans', etc.) that matter the most for your social media strategy. With charts you can visualize and analyze:
- Relations between metrics. E.g. Among influencers of your topic, which gender tends to have the most followers?
- Comparisons of metrics. E.g. Which countries/age groups are talking the most about a given subject?
- Evolutions in time of many interesting social media metrics. E.g. How has the number of followers of a monitored profile evolved over time?
The most common case will be a mix though: a comparison of how different metrics, e.g. number of followers, number of mentions, and so on, have evolved over time. This is where we use segmentation.
(Note that the chart builder is not available in all pricing plans.)
Two Types of Source Data: Mentions Versus Monitored Profiles
A distinction that is important to make early on is that between the two types of source data. This distinction underlies all of CX Social Insights, and it also exists when building a chart with the Chart Builder.
CX Social tracks mentions originating from many different places online, e.g. social networks, forums, blogs, etc. The Inbox lists all these mentions, but we can also count them in many different ways. Each mention has different 'attributes' like country of origin, date posted, etc. Some mentions also have extra attributes specific to their origin. For example, Twitter mentions can have a 'type' which can be tweet/reply/retweet.
Source data type 'mentions' charts are typically used to learn about the number, of a certain subset of mentions that occurred within a certain period, either over the entire period or day-by-day.
- Monitored Profile KPIs
CX Social can also monitor specific social network profiles. From monitored profiles, we then have many other metrics besides just the mentions related to it. For example, from a monitored Twitter profile, we not only have its tweets, but also the number of followers ('users'), Klout score, 'lists' it appears in, etc. So, we can create plots based on these metrics including how they evolve over time as well.
Example 1: Chatter About 2 Luxury Car Brands on Twitter
Let's look at the steps to build a chart explaining what they do along the way. Below is a chart comparing the number of mentions found on Twitter for two different topics, Aston Martin and Lamborghini, over the period of July 6th until July 26th.
Charts comparing the evolution of mentions found over time for different 'segments' like this are a very common use case. In this case, our segments are the topics we want to compare, but you can define meaningful segments in many other ways (e.g. comparing mentions with different tags).
To build this chart we have used the following settings.
This speaks for itself. We named our chart, "Twitter Chatter About 2 Luxury Car Brands."
This is the data you want to 'feed' to your chart.
- Type: There are 2 very fundamental 'types' of charts possible: charts that use mentions as their data and charts that use data from monitored profiles. In this case, we choose the mentions option because that is the 'raw data' with which we want to work.
- Topics: We select 2 topics to use data from: Aston Martin and Lamborghini (to select multiple topics, hold CTRL/CMD while clicking).
- Filter: Within the mentions from our selected topics, we can define a filter and only show data matching that filter. For our chart, we only want Twitter mentions so we create a filter for Source > Twitter.
- Date Range: Similarly, to the date picker found everywhere in CX Social you can choose from a preset date range or define your own. We select custom dates from the drop down and then July 6th - July 26th using the date picker.
Here we define how our chart will visualize our source data.
- Chart Type: You can choose between six different ways to visualize your data, each more suitable for certain types of data.
For showing an evolution over time, a line chart or area chart (the four leftmost types) is most appropriate. For comparing quantities between a "not-too-large" 'set', a column chart or bar chart is most suitable.
(By default, only the upper 4 options are shown. Click 'show all chart types' below them to expose the others.)
Take notice how all bars define a horizontal and a vertical axis, except for the pie chart which will just have a segmentation, i.e. comparing one metric for different 'segments' (e.g. topics, genders, age groups, etc.).
For our chart comparing the evolution of Twitter mentions about the 2 car brands, we choose the first option, line chart.
- Horizontal Axis: The most typical case is to use publish date to display an evolution over time of our metric using a line chart. But, it can also be many other things like country, time, type (e.g. tweets/retweets/replies) where a column chart will typically work best.
- Vertical Axis: The actual metric we want to know about. For charts like ours, where the source data type is mentions, this will typically be total number of mentions. But, if you are looking at Twitter specific charts, you could also look at number of followers (for example, when trying to discover in which countries your influencers have most followers).
- Segmentation: This will cause the different 'lines' in our line chart allowing for visual comparison. For our chart, we want to show a different line for each topic from our source data, so we select 'topic' as segmentation. But, you could just as well show the number of mentions evolved for different tags, different countries, or even different custom filters/searches you define right here.
Note: the sum of the numbers of the different segments might not always add up to the total amount of mentions in a topic because it's possible for a mention to be included in multiple segments, or even not be included at all, all depending on your segmentation.
We are now done defining our chart! You can use the buttons at the bottom to either create your chart, reset your settings, or save your settings as a template for later use.
Press Save Chart As Template to store the settings you created for later reuse. Saved templates will appear as a link on the left side bar, and clicking it will then reload your settings and immediately display the corresponding chart.
Saving a chart template is particularly useful when you choose a 'dynamic' date range, like 'past week' or 'past 28 days'. The next time you view your chart it will show with the settings you like but with up-to-date data! Note that on the sidebar there are also some useful preconstructed templates from which that you can choose.
Example 2: Comparing Followers/Fans of Different Social Network Profiles
Let us now briefly look at a chart of the second source data type, i.e. monitored profile KPI's, for comparison. See the paragraph on source data types at the start of this article for a refresher.
We give fans of a Facebook page and followers of a Twitter account the common name 'users' allowing you to compare these. Let us examine the evolution of these users between the Facebook page and Twitter followers over the past week (as of this publication in August 2012). This is the chart:
Both accounts have increased their follower count in that week, but as you can see, not at a pace that has a great impact. On the vertical axis, you can see that both have around 40k (40,000) followers!
These are the settings used to build this chart:
We select data type, Monitored Profile KPI's, because we are not counting mentions, but instead, a direct property of the profile(s). In the list of monitored profiles, (you can add profiles to monitor in the CX Social settings) we select 2 Twitter accounts.
For defining the chart, we can use all the default selected options, bearing in mind that 'users' means 'followers' in this context:
Given a graph defined using the settings above, you can go a lot further still and customize exactly how things are shown. These options are made visible by pressing the Show Advanced Options button. The advanced options are explained in more detail in this article.