Builds 19 - Call Centres Data Reporting

DavieJoe · May 9, 2022, 1:39pm

True story

BINavPete · May 9, 2022, 2:35pm

Thanks for positive feedback. Much appreciated
Reckon @DavieJoe was taking in the sun in the UK and ran out of time to submit.

Pete

DavieJoe · May 9, 2022, 3:08pm

Scottish people can only handle 10 minutes of 20 degree heat, I had 30mins. Burnt to a crisp!

BrianJ · May 9, 2022, 5:15pm

@DavieJoe ,

Oh no, you don’t. Not this time, my friend. You still have nearly 14 hours before the deadline, so get crackin’.

Don’t make me send @FPastor to your house…

Brian

DavieJoe · May 9, 2022, 5:46pm

I will look to complete it but over the next week or so.

Gustaw · May 9, 2022, 10:29pm

Hi everyone!
How’re you doing guys?

Below you can find my idea for EDNA challenge #19.
Here is a link to interactive published-to-web report:

Below I’ll try to highlight a few interesting features, broken down by given report page.

General info:
All pages have the similar structure and include:
a) Main nagivation pane (left upper corner: summary, monitoring, HR etc.) - main sections of the report
b) Additional navigation (left pane) - refresh / back / home button
c) Main date filter (left pane)
d) Additional text / info (upper section) - info related e.g. with current selections
e) Supplemental filters (right pane)
f) KPI cards - either with YoY change or not - depending on the report page
g) Main sector / space - analysis / visualizations
h) Each page alows to use drill-through analysis (which navigates to either separate “HR performance analysis” or to “CC log”
i) most of KPI cards has applied tooltips that shows both top / bottom 3 agents by given metrics (they can be activated by hovering cursor on specific KPI card)

1st page: Summary
Each KPI header is a button, which enables to “drill through” to specific analysis. In other words, summary report page focuses on 7 different metrics and allows to reveal 7 different perspectives.
Each section is created in very similar way to increase report readability.

2nd page: Monitoring
Main application of this particular report page is to analyze data in daily granularity. On the other hand however, there is a slicer which enables to dynamically change granularity to week/month/quarter/year basis with maintaining hierarchical structure of x-asis at the same time.
There are also switch buttons to quickly change metrics.

3rd page: Day & H Stats
Main application is to analyze given metrics by hour/day. Heat map is additionaly enhanced with formatted background colour & bars.

4th page: HR Perform
This page mainly focuses on analyzing given metrics by manager & agent (or employee).
There’re included three different views (tabular overall, tabular with data in monthly perspective, and small multiples)

5th page: Forecasting
This page enables to make a simplistic forecast (of total charges $ in 2022) based upon “what if scenarios” including variability of predicted number of calls / predicted avg. duration of single call and predicted change in charge value per min. All of this is separately “linked” to different type of call, as they might vary due to different specification.

6th page: CC Log
Generally it’s a form od “database” or log, which contains all of individual calls in case of need for analyzing particular records.

I’ll do my best to share more details along the way.
Meanwhile, congrats for all participants. It’s cool to see such great commitment. It’s look like the number of entrants is much more impressive comparing to recent challenges. Keep up your excelent work

All best to everyone!

JarrettM · May 9, 2022, 10:47pm

Here is my entry for the Challenge. I was able to put this together at the last minute. Going to do my upcoming Summit video on this entry.

Thanks
Jarrett

Josimar_Biosse · May 10, 2022, 2:14am

“I accepted the Enterprise DNA challenge.”

First entry in a challenge and it seemed like a real world scenario. Congratulations to EDNA team and our head of challenges @FPastor for putting this one.

Quick notes:
I’ve struggled a lot since I know nothing about this call center stuff. Had to make some research to understand the industry, the KPIs, etc. Ended up putting just some basic metrics. I think it’s good for a start. Didn’t put much effort on the visualization part (hope to get better in the next visualization summit), but used some tools from the Analyst Hub, like the color palette. It speeded up the basic coloring part.

Also didn’t use many DAX, since I’ve concentrated my efforts on transformations in Power Query. The “Day Shift”, “Night Shift” column/filter was created there as an effort to see if there are any relevant differences calling during day (6 a.m to 6 p.m) or night (6 p.m to 6 a.m).

Realized that manager performance is not depending solely on handling time or number of calls handled, but it mostly depending on what I called “strength in numbers”, since the best performing manager are the ones with a larger team. So, I’d advise Mr Mark Phoney, the CEO, if he wants to increase the revenue (considering total call cost as revenue), then more human resources are needed, mainly for Spokane, WA.

Used the map here as a “natural visual filter”, and used smart narrative to get some quick insights.

It was both challenging and fun this one. This time I had enough courage to press the send button and I hope to get better in time.

All the best

link to the report

JB EDNA Call Center Report

sam.mckay · May 10, 2022, 3:02am

Incredible report! Well done, looks fantastic

Also really love the idea from Tim’s report that I just want to shout out. Great use of dynamic drill down

sam.mckay · May 10, 2022, 3:03am

Fabulous design as usual Gustaw. Stunning looking report

BrianJ · May 10, 2022, 5:10am

All,

Attached please find my submission for Daily Challenge #19. I’m not so much submitting this as I am evicting it from my head, where it is been living rent-free for the past month.

What seemed like a simple dataset at first glance turned into one of the most challenging reports I’ve ever done. Part of this was the nature of the data, part was what I set out as my objectives, and part was due to some dumb choices I made (for example the custom radial chart seemed like a good idea until I realized to get it to work across my navigational structure required setting 300 different button states…).

In terms of objectives, I wanted to focus on the following:

Making the analysis as actionable as possible, centering on specific actions that I thought followed directly from the analysis findings, and including those findings and recommended actions right up front in the report.
Integrating Power BI and R to perform a full statistical analysis to determine which factors likely did and did not have an impact on cost. Because of my focus on making the recommendations actionable, I thought cost was a logical centerpiece for the report.
Continuing to learn how to use some new tools from our recent technology partners – Zebra BI and Numerro.

As I explained in my report, one of the things that made this analysis challenging was that the data were not distributed normally, and thus common statistical tests such as t-tests and analysis of variance were not appropriate. Instead I had to rely on nonparametric statistics that relax the assumption of normality. This presented some unique challenges, particularly in terms of extracting the data dynamically from R scripts run within within Power Query to properly structured tables that could be used dynamically in the subsequent analysis.

I will write this all up in a detailed report this week, but wanted to get this submitted (just,) before the deadline.

Really looking forward to more deeply exploring all the submissions posted this week. Thanks to all for participating, and to Federico and his team for running an awesome challenge.