Skip to content
English
Contact us
  • There are no suggestions because the search field is empty.

Power BI: Influencers Page

The Influencers page uses Power BI's built-in capability to automatically identify which factors significantly influence your KPIs. Unlike the other pages, where you explore the data yourself, this page attempts to bring up what's driving your scores up or down, without you having to go looking for it.

How to use this page

How it works

The Influencers visual uses statistical analysis to identify what makes your selected KPI increase or decrease. It looks at the different ways your data is segmented, by age groups, wards, themes, categories, and so on, and identifies which segments show significantly different performance from the rest.

The key word here is significantly. This visual only surfaces findings it considers statistically meaningful based on your data volume and the size of the difference. If nothing appears, it doesn't mean nothing is happening in your data, it means the visual couldn't confirm anything as statistically significant given your current data and segmentation.

Setting up your analysis

At the top of the page, select which KPI you want to analyse (for example, sentiment or TSM score), then choose whether you want to see what makes it increase or decrease.

For example, selecting "sentiment" and "increase" might reveal that when customers mention staff being helpful or reliable, sentiment rises significantly. That's a useful positive finding worth sharing with your teams.

Reading the results

The page displays a list of themes or segments ranked by their influence on your selected KPI. Each bubble represents a factor. Hover over it to see an explanation, something like "when customers mention X, sentiment is likely to increase by Y." It also shows what percentage of your data this finding is based on.

This percentage is important. A finding based on 0.5% of your data is worth noting, but treat it cautiously. A finding based on 15% of your data is considerably more reliable.

On the right-hand side, a bar chart shows all of your themes ranked by their average sentiment or KPI score. This gives you context for the influences on the left, you can see where each significant factor sits relative to everything else in your data.

If nothing appears

There are a few common reasons the Influencers visual might return no results:

  • Your data is too filtered. If you've narrowed down to a very specific subset, one property type, one ward, one month, there may not be enough data for statistical significance.
  • Segmentation. This visual works best when you have good demographic data uploaded to Wordnerds. The more ways your data can be segmented, the better this visual performs.
  • The variations in your data aren't large enough. If everything in your dataset performs fairly similarly, the visual won't flag anything as a significant influencer.

Key points

  • Purpose: Use Power BI's built-in capability to automatically identify factors that significantly influence KPIs.
  • Statistical analysis: Only shows factors deemed statistically significant, not all patterns in your data.
  • KPI selection: Analyse what makes your chosen metric increase or decrease.
  • Influence ranking: Bubbles sized and ordered by strength of influence on the selected KPI.
  • Data percentage indicator: Shows what proportion of data each finding is based on, check this to assess robustness.
  • Context graph: The right-hand chart shows all themes ranked by KPI for broader context.

Starting point: Use this page to bring up potential findings, then investigate further using other pages.

Use case

Use this page as a starting point rather than a complete answer. If it tells you that mentions of "emotional effort" significantly decrease a KPI, that's worth following up on. You'll want to look at it on another page, such as Category and Theme Based Frameworks, to understand what customers are actually saying about emotional effort and what the crossover themes are, so you can put preventative actions in place.

Think of the Influencers page as a quick screening tool that might highlight something you hadn't noticed. It can bring up the what; the other analysis pages help you understand the why and decide what to do about it.

Zoe-1    ✍️ Article written by: Zoe, Customer Success Manager 

Still in need of some help? Give us an email on support@wordnerds.ai or reach out to your CSM directly.