Over the last few weeks Nintex has been talking about and showing a new Enterprise service called Nintex Hawkeye. Nintex Hawkeye gives you an unprecedented view into your process portfolio enabling you to monitor your business, get answers quickly and make informed decisions. Businesses can use Nintex Hawkeye to identify which processes yield the best business results and target those that are ripe for improvement. Nintex Hawkeye brings a new generation of analytics for human-centric processes, the people who use them, and the content they create.
A key aspect to this capability is the Nintex Hawkeye Process Intelligence lens. The Process could be 1 or more workflows across multiple system, that contribute to a single "Process". With the Process Intelligence Lens, we want to dive into the information, not just about the workflow/s but also the context information that is involved as well. The first thing you need to do to get the ball rolling is to decorate the workflow. Now by decorate I mean use the new Nintex Hawkeye beacon actions inside the workflow to gather the information I want more insight from.
To illustrate this process I'm going to build it out.
I'm automating a process for a hospital. The process is to gather bedside patient surveys on how they were cared for, did they observe anything like doctors not washing their hands, or nursing staff not explaining what they were doing etc. This information needs to be gathered up and rolled into a weekly report, but in addition to that, it can be used to see trends in behaviors. For example, nursing staff in a particular ward or department are always receiving top marks for being courteous, when the rest of the hospital is neutral. Collecting this data and being able to track and visualize these key metrics over time provides management with the necessary tools to make informed decisions.
I've built out my little workflow to processes the information for the survey. Traditionally is was taken from a clipboard and added to an access database, which would push it to a SQL table. My workflow was pushing the data to the same database. However now that I have Nintex Hawkeye I don't really even need to do that. In my workflow there is a task I'd like to measure. This is something I haven't been able to do before. So using the Nintex Hawkeye beacons:
I can use the "Beginning Beacon" and the "Ending Beacon" to not only gather information but to automatically measure the time it took between the 2 beacons.
Configuring the 2 different action is relatively similar except for 1 thing. In the "Beginning Beacon" you specify what your stage is called:
and in the "Ending Beacon" you simply select which stage to you want it to connect to and wrap up.
You will also see in the above beacons that I have added gather points or "Beacon". Using the single action I can gather a whole lot of information.
The third beacon is essentially a stand alone beacon, which allows you to simply push information to the Nintex Hawkeye service at a designated point in the workflow.
With my workflow, it is still being retro fitted so I'm pushing the survey data to the SQL DB and I'm also queuing it up to push to Nintex Hawkeye.
In this beacon I'm gathering all of the survey data and queuing it up as well.
Ok for this example we are going to keep it simple. After publishing my workflow and having some users use it, I can see the items queue and then shoot off to the service. You can get to this page in central Admin
I'm ready to take on the next phase of the "Process Intelligence Lens"!
Log into your Nintex Hawkeye tenant and select the Lenses link. Select "Create Lens"
and then select "Process Intelligence Lens".
First things first, give the lens a name and description.
Once you add the workflow, it will appear on the right hand side and display a progress bar for extracting the data. Once that extraction has completed you are go to go.
The next step is to take a look at the "Out of the Box" dashboard for the process Lens. Here we get a pretty generic look at the data for this process. I can see how many executions, duration, tasks, participants and ROI calculations.
But I really want to start adding some custom graphs based on the data my beacons are collecting. In order to do this, I can do a few different things. I can get the OData feeds and consume them in Excel or a reporting / BI product. However for this example I'm going to get the PowerBi template and open it in the free Microsoft Power BI Desktop product.
Once the template has opened, I need to connect my data source to it.. PowerBi will prompt me for this as soon as the template opens
Paste the OData link that you can get from the same window as the template download. It will look a little like this
After you select load, you will be asked for your credentials. Select Basic and enter the credentials you used to log into the Nintex Hawkeye Portal.
Alright you still with me ? This may seem like a long process, but when you are actually doing it, it is fairly painless and short.
Now we are in PowerBi, we have a direct line into the OData feed from Nintex Hawkeye.. Time to build a graph, but firstly lets take a quick look at the data..
Select the 'Edit Queries' button in the toolbar. Over in the left hand menu we have a reference list of all the queries, these are what we want to look at.. In particular, the ones under 'Business Data Model' ProcessStages, and ProcessBeacons.
Selecting the ProcessStages query, I'm going to see the stage information from the Beginning and Ending Beacon
Looking at the beacons there is some particular data I want. Because all the beacon data is bundled up together in a single query I need to strip it out to make it easier when it comes time to visualize it. First of all, because I want to keep it neat, I'm going to create a new Group by right clicking in the query area and I"m going to create a reference query on the ProcessBeacons query. You do this by right clicking on the query you wish to reference. I'm also going to rename this query to 'HospitalRecommend'
Now that I have my reference query I can add steps to this without effecting the original, referenced query. So what I want to do is just show the data that is relevant to see if patients would recommend the hospital to their family and friends.
Putting this data to use now, close and apply the Query Editor. In the right hand toolbar I have a new query available called 'HospitalRecommend'.
All I need to do now is select the graph I want and the Value and Name fields to populate the graph.
This is a very simple application of the service, but as you can probably imagine there are hundreds of ways you can slice and present the data.
My biggest word of advice is to seek guidance from the process stakeholders and plan what you want to present well before you start building anything.
Until next time.. Enjoy sinking your teeth into something new..