What if model

All these KPIs are created in the backend depending on the available data and made available as per the audience. Another thing to note is these metrics are benchmark metrics, meaning they compare year to date vs last year to date as shown in the percentage change. Yellow — This section shows a map view of revenue this year vs last. Now there are two features to this: one is after looking at the state numbers you can click on that state or even select multiple states to drill down to another map which shows data at city and then at the zip code level.

The second feature is you can embed an external dataset in this dashboard and overlay it on top of your sales data to derive insights.

Here, for example, we are using US census data white-collar occupation to see how your company is doing across segments, something that your company does not have. Teal — This section helps keep a track of what is the buying behavior, what kind of products are being bought this year vs the last year.

The heatmap or treemap shows revenue in terms of color and number of orders in terms of size. You can click on the product category to deep dive into the product SKU level. Green — This is a revenue trend line graph useful to find patterns across multiple years. You can again drill down to look at trends at the day level and see how sales performed during sale events and after major campaigns.

In conclusion : Building Analytics dashboards have become a business necessity but designing a BI board should follow certain rules. Care must also be taken to ensure the board does not become complex. BI dashboards can be used to establish a relationship between two or more variables, or to compare two or more variables alongside, to name a few.

Oyster is not just a customer data platform CDP. At its core is your customer. Explore More. Your email address will not be published. Save my name, email, and website in this browser for the next time I comment.

From time to time, we would like to contact you about our products and services, as well as other content that may be of interest to you. By ticking on the box, you have deemed to have given your consent to us contacting you either by electronic mail or otherwise, for this purpose.

Customer Data Platform. Vinay Dabhade 01 Apr Very quickly, data dashboards can be divided into three broad classes: Operational Dashboard: This type of dashboard allows users to conduct data-driven actions. Benefits Of Analytics Dashboards The function of any business intelligence dashboard is to extract value from the data collated, also known as key performance indicators KPIs.

Here are some of the benefits of implementing dashboards in your business operations: Save time Before dashboards, there were cumbersome sheets and charts which took a lot of time to draw up. Better forecasting When a business has more insights into the buying cycle of customers, future demand can be better predicted. Last post by The Rat in Re: "quotBlack widow"quot unli Last post by zenrat in Re: Messerschultz Me Last post by DogfighterZen in Re: the one with a littl Last post by Old Wombat in Re: Chris builds yet ano Last post by su27rules in Re: Whiffie Awards Last post by joncarrfarrelly in Re: Dramatic but Overrat No New Posts Redirect Board.

What if - Info Center. Recent Posts. What if - Recent Posts. Subscribe to Webslice. Forum Stats. Users Online. SMF 2. Tips, Tools, and Techniques That may be of use to other members. Post them here. What-if related Products For the discussion of products that will be of interest to forum members. Upcoming events Use this thread to coordinate get-togethers on upcoming contests, events, conventions, etc. Aircraft, Armor, Weapons and Ships by Topic Please do not create a new thread if there is a thread about a particular subject open.

Let's look at how to build what-if scenario analyses in two different tools; Excel and Causal. Excel features a section of 'what-if' tools, to help users understand questions like those posed in the sections above. To use scenario manager, you first need to build up a model. This isn't as difficult as it might sound, a model is simply a set of inputs, with an output that's a function of those inputs.

For example, we might create a model with the inputs Price of Bread and Bread Sold , and the output as Revenue :. Price of Bread can be whatever you want it to be, Bread Sold should be a function of Price of Bread , and revenue should be the last two numbers multiplied together. Once you've done this, you can go ahead and open up Scenario Manager, and build your scenarios by clicking 'Add Once you've created your scenarios, you can hit 'Show' to see how the output of your model revenue differs between the scenarios you've defined:.

Excel's scenario manager is a good tool for analysing models that are already built in Excel, but it has one large downside. If you're creating complex models and moving cells around, this can often break your scenarios.

This happens because scenario manager's Changing Cells are fixed, and won't respond to changes in your model. If the above felt a little clunky then don't worry, what-if scenarios are much easier to build in Causal.

We'll start by building our model. With the plugin, you can perform inference on a large set of examples and immediately visualize the results in a variety of ways. Additionally, examples can be edited manually or programmatically and re-run through the model in order to see the results of the changes.

It contains tooling for investigating model performance and fairness over subsets of a dataset. The purpose of the tool is to give people a simple, intuitive, and powerful way to explore and investigate trained ML models through a visual interface with absolutely no code required. The tool can be accessed through TensorBoard or directly in a Jupyter or Colab notebook. For more in-depth details, demos, walkthroughs, and information specific to using WIT in notebook mode, see the What-If Tool website.

When opening the What-If Tool dashboard in TensorBoard, you will see a setup screen where you provide the host and port of the model server, the name of the model being served, the type of model, and the path to the TFRecords file to load. After filling this information out and clicking "Accept", WIT will load the dataset and run inference with the model, displaying the results.

For details on the different features of WIT and how they can aid in model understanding and fairness investigations, see the walkthrough on the What-If Tool website.