[MUSIC] Welcome back. In our last lesson we took a look at a case using the World Indicators data set where I elaborated a bit on the first three design principles. Choose metrics that matter, keep it visual, and make it interactive. For this lesson we're going to continue using that same visualization, using the World Indicators data set to look at the remaining of those best practices. After this lesson you will be able to assess how to keep your data up-to-date, design your visualization with simplicity in mind for access in use, and discover the seven mistakes to avoid. We have a lot to cover, so let's get started. The fourth best practice tableau offers for creating dashboards is to keep the data up to date. Look again at this visualization. It shows the years from 2000 to 2012. The intention of this visualization is to give a historical snapshot. So let's be clear about that. But of course it's 2017, and given that the most recent set of data are from 2012 it's likely that the data were collected and reported on in 2011. So the data are pretty out of date. In particular, if you think about what happened in the world since 2011 you know that there's been a lot of conflict. For example, there has been absolutely terrible and horrific civil war in Syria since 2011. Yet If I go ahead and show the data on Syria here You can see something very interesting, and that is the infant mortality rate is falling. So I'm just going to draw a really quick trend line, and you can just follow along here. We've shown you trend lines previously. And just putting a trend line over both of these and you can see here what I'm saying is that the health expenditure as a percentage of GDP is falling, but you can see the infant mortality rate is falling as well. And unfortunately, in a civil war, that's not the case. There's been a tax on hospitals and other healthcare centers. So, even though it seems as though there's a clear trend towards lower infant mortality we know it's actually not the case. And so, I'm just going to get rid of these trend lines, don't mind me here while I keep talking. Presenting data that is not up to date can skew the message to give a false impression. Therefore, it's absolutely imperative to have current data, when appropriate. In this case, the visualization fails in that metric. By the way, I'm saying openly that this visualization fails in this metric, even though I created the visualization. I think it's okay to be critical of it. I want you have a critical eye when you're designing visualizations and dashboards to ensure that you're giving the best visualization that you can give to the world to use. It's also essential to be upfront with the user and audience about the limitations of the data and conclusions. Best practices number five is make it simple to access and use. This is easily accomplished by posting this onto Tableau Public. This particular visualization I can post to Tableau Public, and it will be no big deal because the data are public already. However, there are lots of visualizations that you do not want post on Tableau Public. Why? As we've talked about back in course one, all underlying data are freely available when you post to Tableau Public. So if you post it to Tableau Public, make sure that you're okay with your data being shared. If you're not, don't post it to Tableau Public. If there were a reason not to post to Tableau Public, how else could you make it simple to access a news? We could share the file, and have people download Tableau reader. We could find money to buy Tableau online licenses. Or if we're really lucky our firm might have Tableau server. Once you have that, the whole other world opens up for you. With Tableau server, you could store proprietary data, and use Tableau because it's hosted on your organization or company's own server. Finally, best practices number six is mistakes to avoid. Tableau lists seven mistakes to avoid. And I definitely agree with all of them. And I'll see if my little visualization avoids these mistakes or not. The first is starting off with too much complexity. We could see in our visualization, it isn't very complex. It could use more labeling to make it a little bit easier to read. But, you can see what health expenditures as a percentage of GDP are, and what infant mortality rates are. There's a legend and things like that. And it's very easy to read, it's not very complex. A line graph people understand because of one of the gestalt principles that we learned about in course two. The second mistake to avoid is using metrics that no one understands. In this case again, these are very straight forward. We know what infant mortality is and we know the rate. We know the held expenditures as a percentage of GDP. There's information there that we could look up if we needed to, or we could provide the definitions in a dashboard to ensure that everyone is on the same page about what the metrics actually mean. The third mistake to avoid is decluttering the dashboard with unimportant graphics. In other words, what I mean is that, don't clutter the dashboard with unimportant graphics, but declutter. If you clutter it with unimportant graphics, your audience is going to be putting their brain to work on those an important graphics. So avoid them all costs, they definitely are not in my dashboard, or rather visualization. The fourth mistake to avoid is waiting for complex technology and big deployment projects for business intelligence to happen. We talked a lot about business intelligence back in course one, so you can refer back to that if you need to. These deployments take a long time and I tell you, the temptation is to try to wait for that because once it happens, we're going to say, it's going to be so much nicer. Once we have Tableau server everything is going to be great or once we have Cognos everything is going to be great. It's not trivial to set these things up, it could be years, and you do yourself a disservice by waiting around for those things. That being said it's often frustrating because you know that there are better solutions out there, but you just can't get to them yet.
