This has turned out to be an epic weekend.  The Summer Games are wrapping up, NFL preseason football has started, and what did I happen catch this evening after coming inside from the 85 degree heat wave here in Seattle?  That’s right, Shark Week has kicked off over on the Discovery Channel.  If there is a better trifecta out there, I’m going to be hard pressed to find it.

I’ve always been interested in sharks so naturally one of the first things that came to mind this year was whether I could hunt down (I guess that pun was intended) some data on shark attacks.  There are a couple of places where you can find some information, but finding pristine data here doesn’t look possible.  Nonetheless, if you head over to the Global Shark Attack File, you can get to an incident log that you can download in  Excel.  Once we’ve got the data in Excel, you are not too far away from being able to display it on map with Excel 2013.

The Shark Attack File contains data about locations, but as you will see here in a bit, the data is far from perfect.  I don’t have exact coordinates and as you can see from a snapshot of the data, if you wanted this to be perfect, you would have some work in front of you.

That said, one of the features of the maps in Power View is that the data can be automatically geocoded by Bing.  This means that even for cases where the data is not absolutely perfect (think about the times where you’ve used Bing to get directions and didn’t spell something correctly and it still was able to resolve your route), it still gives it a best shot effort.  The nice thing about having Power View in Excel though, is that if you notice the data is not being plotted on the map correctly, you can just go in and fix the data and the spreadsheet and your map will be updated.

Let’s get started.  I’ve imported the data into Excel as you can see here.  The data goes way back to even dates before 1554 where apparently it was a good idea to be wearing armor in the event that you were bitten by shark.  

Next, we go to the ribbon in Excel and Insert –> Power View.  This recreates the table in the Power View sheet that is generated.  Go ahead and delete that table.  Now, go to the model explorer over on the right and add Country and Case ID to the table.  Then, from the drop zone, select the Case ID field and switch it to a count.

Next, go to the Design tab of the ribbon and choose the option to change the table into a map.

And there you have it.  By the way, there is another new feature in Power View that allows you to add hyperlinks to your report.  Since I pulled the Excel file down from the Shark Attack File, I’ll place a link to this on the report.  To do that, simply insert a text box and type in the text along with the URL.  Here is the textbox I added to the report.

Finally, I want to drill into the actual locations where the shark attacks occurred instead of just looking at the Country level of data.  To do this, I’m going to flip over to the Power Pivot window and add a couple of calculated fields.  I added one to represent “Country and Area” and then I added another one for “Location and Area”.  This will allow us to drill down a couple levels.  Mark these two fields as Geographic Type = Place, which you can do rom from the Advanced section of Power Pivot. 

Switch back to Excel and the Power View sheet and you should see your new calculated fields.  Add them to the Locations drop zone.

  Now you can drill down (double click) on the Countries to see the states and the locations of the attacks.  Here I’ve drilled into California.

If you want to see which of these attacks were fatal, simply click on the Fatal Field. 

You’ll notice here that the data is not perfect, as some of the attacks look like they occurred many miles away from the shore (maybe there are aquariums in the area), but if you hover over those data points in question you’ll see that the location information for those points is just not defined well enough.    Still, for five minutes of your time with a data set that is far from “clean”, this definitely gives you a very good idea of where shark attacks are occurring. 

In future Shark Week related posts, I’ll add in a few more pieces of functionality, but for now, they are starting to show sharks jumping out of the water again on the television and I’m not going to miss that :)

Download this workbook here.

-Sean

Once you go through the process of getting the app installed it shows up in the ribbon here:

This inserts what is referred to as a task pane app which shows Live Medal Counts for the Olympics.

While this is a really nice presentation of the data in the task pane, the first thing that caught my eye were those bright buttons that read “Add to Spreadsheet”.  Are you thinking what I’m thinking?  If you click on that button, the data does get added to your spreadsheet like so.  And once you have data in the spreadsheet, you are just one click away from being able to do a couple of other things:

• Insert the data into PowerPivot
• Create a Power View report in Excel

Technically, creating a Power View report in Excel will also insert the data into PowerPivot.  Yes, you read that right.  When you create a Power View report in Excel by selecting data from a table or a range, behind the scenes a PowerPivot model is generated in the workbook. 

Here’s the spreadsheet with the data added.

One great piece of interactivity here worth noting.  When you click on a row in the table in the spreadsheet, the task pane updates to show stats for just that country.  Pretty cool.

If you go to the ribbon with the PowerPivot add-in enabled, you can click on the option to add this table to the data model.  As I mentioned above, if you want to go straight to PowerView before doing this, you can do that as well and it will create the same table in the model.   Once in the model, you can set the model table to update whenever the data in the spreadsheet table updates.  Now if you look at the app in the task pane again, notice it has the option to “refresh the data”.  So you can add this live medal data to a spreadsheet, add it to PowerPivot, hit refresh in the Medal Tracker app and now that data is up to date in PowerPivot and any Power View reports you create as well.

The Medal Tracker app doesn’t just have this one table of data though.  It contains a whole set of data feeds for medal history, sports lists, athletes, etc.  Pretty much everything we need to build out a complete PowerPivot model for the Olympics.  We can add all of those tables to individual sheets, add them to the model, relate them together and we are well on our way to being able to make some interesting views of the data. 

Fast forward through adding all of the data tables and let’s take a look at today’s Power View feature, drillable visualizations.  After creating the model based on these tables, add a Power View sheet that shows the Count of Medals by Discipline and Event and convert the visual to a Matrix.  Both of these fields are part of a hierarchy that is created in the model, but in Power View you can drill down on any field that you like by simply adding the field to a drop zone in the field well that accepts more than a single field.  In our example right now, the drop zones look like this:

To enable drill, all you need is to do is to go to the ribbon and choose the option Design –> Show Levels –> Rows – Enable Drill Down One Level at a Time.

You can now drill up and down on disciplines.

Convert the matrix to a column chart and the drill feature is carried over.  You can double-click on the columns in the chart to drill down to the next level. 

When you want to drill up for the chart, the entry point is in the “floatie” area right above the chart next to the option that opens up the filter pane.

Lastly, you are not confined to just fields that are part of a hierarchy for drill.  For example, you can add Team to the top of the drill path here and start the drill down from team instead. 

Now I’ve got a bit of conundrum.  In my previous posts I built up a workbook based on a different set of data, but the opportunity to check out the Medal Tracker app for Office was just too good to pass up and not include here.  As a result,  I have two separate workbooks based on Olympics data.  I’m in the process of consolidating the two and I hope that by my next post, I’ll have that done and we’ll continue the tour based on a single workbook. 

If you would like to download the workbook and try it out, I’ve placed it here.  I’d also recommend that you check out Dan English’s latest blog entry on Analyzing SQL Server Saturday data with Power View over at http://denglishbi.wordpress.com/2012/08/06/sql-saturday-event-analysis-using-power-view-and-office-2013/ 

-Sean

Welcome to Day 4 of this series on Visualizing the Olympics with Power View in Excel 2013. In yesterday’s post we looked at how you can add embedded images to the PowerPivot model that is present in Excel whenever you are using Power View content. 

Worth noting on its own, The PowerPivot in-memory engine (also known as xVelocity) is now fully integrated into Excel with Excel 2013.  As a result, whenever you create a Power View report in Excel or whenever you create a PivotTable that is based on multiple tables in Excel (a very cool new feature in Excel if I may add), behind the scenes a PowerPivot model is generated in the workbook. 

For today’s post, we are going to look at how you can save your workbooks to Office 365 and then interact with them in desktop browsers.  To take advantage of this capability when you sign up for the Office Consumer Preview, choose the Office 365 Enterprise offering.

Once you sign up for Office 365 and download the Excel 2013 Consumer Preview, you will then be able to create a Sharepoint site such as the one displayed below:

As you can see, I now have a document library on my team site here in Office 365.  You can simply drag and drop your workbook to the document library, but once it is there, all you have to do is click on it and it opens in the browser using the Microsoft Excel Web App Preview.

The workbook is now being displayed in the browser and our Power View content is fully interactive.  I can continue to change the sport I want to look at on the map by simply clicking on one of the images that we set up in yesterday’s post.   The filter pane is also available. 

Right now there are a couple of restrictions that you will need to be aware of in the Office 365 setting.  The first is that all of the data must be in the internal Power Pivot model that is part of the workbook.  While it is possible to connect directly to Analysis Services Tabular Models that are deployed on an Analysis Services server, this feature is not yet available in the Office 365 setting.  The second limitation right now is that the workbooks must be 5 10 MB in size or less.  While that may not seem like a lot of data, we have examples running where we have managed to store more than one million records in PowerPivot that fit within this size restriction. 

Welcome to Day 3 of our effort to visualize the Olympics using Power View in Excel.  In yesterday’s post, we looked at Tiles, pie charts in maps, and dabbled a bit in the filter pane.  Today, we are going to introduce the PowerPivot add-in and one of the features of the modeling experience in PowerPivot that will allow you to add images to your reports.

First things first.  You need to download the Excel 2013 Consumer Preview from here.  Once you do that, you can download the resulting workbook from today’s exercise here. 

Now that you have the workbook, the next thing you will want to do is enable the PowerPivot add-in if it is not already enabled.  To do that, simply go to File –> Options –> Add-ins –> COM Add-ins and

then make sure the Microsoft Office PowerPivot for Excel 2013 add-in is enabled.

Back in Excel, we can launch the PowerPivot window from the ribbon.  Go to PowerPivot –> Manage.  Once PowerPivot is launched, you will see a set of six tables.  One for Country, Sport, Medal Images, Countries, Sports, and Athletes.  Three of these tables are linked tables and two of these tables came from an external database.   The three linked tables are just like the tables that we have been working on for the last two sessions.  They are simply links to the Excel tables that are in the workbook.   The external tables, on the other hand, are based on a query to a SQL Server database set up.  So why did I do that?  Well, that is the focus of today’s session.

Let’s focus on the Sport Table.  Here you will notice that there is a field called “SportImage” that contains Binary Data.

These are the actual images that are the pictograms that represent the sport.  In order to take advantage of these images in Power View, go to the ribbon in PowerPivot and ensure that Advanced Mode is enabled. 

Once Advanced Mode is enabled, click on the Advanced Tab and you should see in the Reporting Categories chunk of the ribbon that the Data Category is set to “Image”.  This is the property that allows Power View to display the contents of this field as an image.

So how do we get this data into this table in the first place?  First we need a set of images, which I have stored on my local hard drive and then we need a SQL Server script to loop through a table that contains the name of the sports and then does an import of the image into SQL Server.  Below is the syntax I used for this operation. 

Now that I have the data in SQL Server, I can choose to import that data into PowerPivot.  To do that I just walked through the PowerPivot import data from SQL Server option and chose the option to import all the data from this table. 

For those of you who already familiar with the ability to use images in Power View, you may be asking why did I decide to actually import them into the model instead of simply referring to an external image URL that could also work?  There are a couple of reasons for why I opted to actually embed the images.  First, once the images are in the model in the workbook, I have a very portable solution.  I can hand this workbook to someone else and I don’t have to worry about whether or not they will be able to see the images.  The second reason is that if I plan to use this workbook in Sharepoint or Office 365 based scenarios (by the way, you can try out the Office 365 scenario with the Consumer Preview), the Office 365 scenario does not support access to external images. 

With our images now in the Power Pivot model, we can return to Power View and make a change to the report we built yesterday.  When we left the report yesterday, it contained a Tiles container with a map in it.  To add the images to the report, you can simply switch the current Tile By field from Sport Name to Sports Image. That will give you the following map.  

Tiles are not the only place where you can add these images.  In future posts, we’ll cover additional ways that you can use them in your reports.  Note that I have also inserted into the workbook images to support displaying country flags and medals.  The process to upload those into the model was exactly the same.  See you tomorrow!

For the second exercise, we are going to pick up where we left off from yesterday.  You can download this starting workbook here.

First, select the map and then let’s click on the Gender category from the field list.  As soon as you do this, you will notice that the bubbles on the map will turn into pie charts displaying the percentage of men and women participating for each country. 

Notice also that when you click on the entries in the legend the slices of the pie chart highlight that represent that entry.

As an example, clicking on “M” highlights all of the slices that represent male athlete  participation.  Here I’ve also zoomed into the map to center the map in Europe.

Now let’s break this data out by different sports.  To do this, place the “Sports Name” field on Tiles Drop Zone.  Now you can flip through versions of this map by each sport.

Once you place the map inside of the Tiles container, there is additional option that becomes available.  This option allows you to synchronize the bubble sizes across all of the map tiles.  As a result, the sports that have more athletes participating in them will have the largest pie charts across all of the map instances.  If you decide to leave the map in it’s default state, each map will compute it’s own maximum pie chart size.  To set this option, simply select the map and go to the Layout section of the ribbon.  There you will find the Synchronize Bubbles Option where you can set them to be aligned across all tiles instances.

Go ahead and click through some of the sports.  You will find some interesting data points. 

In tomorrow’s post, we will start to work in the PowerPivot add-in so we can add some images to this report.

For starters, the first thing you will need is the Office 2013 Consumer Preview which you can download at http://www.microsoft.com/office/preview/en.  The preview is completely free  and you can install it side by side with your existing Office installation.  As we work through the solution over the next few weeks, I’ll post each day’s workbook here in a SkyDrive folder called Power View Olympics.  I’ll also include a short video that demonstrates the day’s solution as well.

Alright, let’s get started.  For this first example, we are going to start with the list of the 10,960 athletes that are participating in the Olympics and create a map that displays the count of athletes from each country.   It’s in a table in Excel.

Next, select any cell in the table and then go to the ribbon and choose the option to insert a Power View report.

This will create a Power View sheet in Excel that contains a copy of the table. 

If you are interested in looking at some of the athletes and where they are from you can sort the data using the column header text or filter out rows using the filter pane to the right.  However, in our case we are going to create a map, so we will delete this table and leave ourselves with a blank canvas.  Let’s also add a title to the top of the report.

From the field list now, simply expand the Athletes entity and select country.  Your report should now look like this.

To create a map, simply go to the ribbon and choose the Map option from the Switch Visualization Ribbon chunk.

When you choose this option the list of countries will turn into a map with bubbles that are all of the same size as shown below.

A couple of things to note about the map.  The first is that you must be online and have connectivity to Bing in order to use the feature.  Both the background for the map and the individual points on the map are determined by passing your data over a secure web connection to Bing, where Bing geocodes the data and then returns it back to Excel.

Look at that.  We are almost done.  To show the count of athletes in each country, we simply just need to drag the “Name” field from the field list down to the Size drop zone in the field well and your map will look like this.

And with that, we have created our first map for the Olympics.  While you are here you can zoom in and pan around on the map as well as hover over the data points to see the values for each country.

Here is a link to a video showing all of the steps required: 

Power View in Excel 2103 Olympics Data - Day 1

Thanks for following along with this first post.  While there is a lot more functionality available in the map, much of it we will likely cover in future posts, I thought it would be fun to start with a simple exercise for the first post to show you how easy it can be to build a map in Excel 2013 with Power View.

• SQL Server Reporting Services Team Blog Announces Project Crescent Availability
• Project Crescent Overview and Wiki
• Project Crescent: Demos Galore  (session from TechEd North America 2011)

Also, today, we released a lengthy 14 minute video titled Project Crescent: Getting Started and Getting the Most from Project Crescent.  It covers much of the functionality available in CTP3.

In addition to these resources I’m planning on starting a video series accompanied by a set of PowerPivot workbooks.  These videos will be based on real data, all of which I plan to make available for everyone to download.  I certainly view these as works in progress, and as I look at topics to cover, I’ll definitely be making changes to the workbooks and republishing them.  And if you are watching these videos, and you questions or have an idea about how to improve the workbooks with examples, please do feel free to contact me here on the blog.  I’d like this to be as much of a community effort as possible.

This first video demonstrates a feature in Project Crescent that is known as highlighting.  If you want to read a great article on this topic, I highly recommend Stephen Few’s article Coordinated Highlighting in Context.  I’m going to demonstrate the Crescent functionality over a set of maybe two or three videos, and this first one covers the interactivity associated with column charts and column charts that have series and a legend.  The data being used for this demonstration is from the CARS (Car Allowance Rebate System) program that ran here in the United States during 2009. You may have heard of this program by another name, the “Cash for Clunkers” program.   At any rate, being a government program, the data is all freely available for download on the site and I have turned it into a very basic PowerPivot workbook that you can download here if you like.  There are quite a few interesting stories to tell about this data, and I hope to show you some of them.   Over time as I create more videos and demos covering additional topics, I plan to update the workbook and continue to make it available.  Below is the first video on this topic.

Project Crescent Demo: Introduction to Highlighting

To summarize the video, everything in a Crescent report is interactive (except maybe the textboxes : ) ).  People can simply click on a column in a chart and use that as means to highlight data throughout the report.  The other thing you hopefully noticed is that as a report author, I didn’t have to configure this feature at all, it just works.  There are no property pages that open up and require you to be an expert in how the system works underneath to make it all happen.  Much of it happens due to the relationships that are already defined in the model, but as someone who is creating the report or interacting with the data, you don’t even need to be aware of that.  And highlighting isn’t the only thing that happens on the canvas.  Charts can also filter and we’ll look at that feature in a future video. 

In this scenario, we’ll look at an example where you might have a list of stores with just a store name field and you’d like to show their sales and profitability.  The dataset looks like this.

Once you have the dataset, simply add a map to the report by right-clicking –> Insert Map.  This will place a blank map on the report.

Next, select a map and add a tile layer.  This will place a Bing! maps tile layer background on the map.  In our case, we are going to assume that our stores are all in the state of Washington so we need to zoom the map into the Washington state area and we can do that by just panning and zooming the map in the designer.   After zooming and panning the map will look something like this.

Now that we have the state, we can add embedded points into the map.  For our case, we are going to assume that our three stores are in Seattle, Tacoma, and Olympia.  To add the points, first add a point layer to the map using the fly-out.

Now there is a point layer on the map.  Next, select the drop-down on the point layer and choose the option to “Add Point” as shown below.   Add the point over the Seattle city area.

Once the point is on the map, you can select it and bring up its property page.  Give the point a name, and you can optionally choose to open up a lot of other properties for the point using the checkbox override.   For now, we’ll leave this box in its unchecked state.

Repeat this procedure to add Store 2 and Store 3 in Tacoma and Olympia. 

Once the map has all three points added, select the Point Layer in the fly-out and choose the option to bring up the layer wizard.  This will bring up the layer wizard for the point layer.

Select the Bubble Map option and hit “Next”.  On the next page where the wizard asks you to select an analytical dataset, select the dataset you created based on the query at the beginning of this post, and then select “Next”.

On the next page you will be asked to bind the analytical data to the name property we created on the embedded points.  Here’s a screenshot of what that should look like.

On the next page of the wizard, we’ll choose the options for setting the bubble size and bubble color.  In our case, we want to make the bubble sizes dependent on sales and the bubble color to be dependent on profit margin.  Below is what the wizard should look like once it’s filled in.

Click “Finish” to leave the wizard and return to the report design surface.

Before we run the report, there’s one last bit of cleanup we need to do.  Select the Point Layer and bring up the Point Color rule property page.  We want to remove the point colors showing up in the legend on the right.  Select the Legend tab and from the “Show in Legend” drop-down, de-select Legend1 as shown below.

Exit the property page and then return to the report.  Go ahead and run it now and you should see the following:

That’s it. Of course,  you could go in and add a Legend Title, or even change the Point color rule properties to create as many color buckets as you like to customize the report further.