Navigating the platform

This section of the help centre will assist you in learning what each part of the toolkit does, and how to navigate across the platform.

IVP.ai is navigable via the left-hand side panel. Below are the primary features that will be covered in more detail later on in this guide.


Home - This part of the platform will offer access to the other features, an overview of the Workspaces and store any favourited dashboards.


Connections - In connections, we can establish a flow of data into the platform. IVP.ai offers live API connections to the most frequently used third party data sources, as well as the option to upload static files such as Excel spreadsheets or Json files.


Once a connection has been established to a live source, (completed via logging into the 3rd party data source via the relevant button) the data will flow according to the API rules set by the 3rd party and, once this step has been completed, will typically not need setting up again.


If you do not see a specific connector on the connections tab, contact us directly to arrange setting one up.


Datasets - Acting as the platform’s data warehouse, Datasets will store any datasets that have been pulled into the platform. Each dataset will have a name, a creation/refresh date and a symbol in the bottom left of the box indicating one of two things. 


The connections symbol represents a dataset that has been pulled directly from the source. A Pipeline symbol will be present where the dataset is the product of a transformation that has taken place in Pipelines, and has been toggled on to save.


We can also perform a preliminary analysis by exploring the dataset at this stage, which will open up the dataset so that we can see & search each individual row, summaries for the columns, correlations & distributions, and the Advanced Explorer which acts similarly to a pivot table.


Pipelines - This is where IVP.ai enables data transformations to clean datasets and prepare them for analysis. Typically, data transformation is the most complex step in any data analysis process, so depending on your use case, saving a pipeline may be a good idea if you are wanting to run the same type of analysis again and again.


Pipelines features a drag & drop interface where we can transform an original dataset through any of the following actions (these will be explored in more detail on the Pipelines help centre tab): 


  • Join
  • Append
  • Aggregate
  • Calculate
  • Filter
  • Tag
  • Remove columns
  • Rename columns
  • Date time extraction
  • Convert 0 to null
  • Convert null to 0
  • Pivot

Pipelines also offer the option to harness industry-leading AI models. IVP currently offers four types of AI model including:



  • Forecasting (predicting future values)
  • Grouping (clustering variables with other similar variables)
  • Missing values (predicting values by learning from training data with known values)
  • Reduce complexity (a form of dimensionality reduction, to reduce noise from your dataset to account for outliers)

Analytics - This is the visualisation suite of IVP.ai. Here, you can view the output of any given dataset in a corresponding graph/chart format. IVP.ai will not allow an incompatible dataset & analysis format e.g. a time series chart plotted from a dataset without chronological columning.


When creating an analysis for the first time, you will have the option to choose a template analysis from:


  • Time series
  • Forecast
  • Describe
  • KPI Dashboard
  • Category
  • Outlier
  • Geo-spatial

We will cover these template analyses in further detail later on in the guide. You will also have the option to create a custom dashboard with a mix of the above analysis templates, in addition to other customisation features.


AI Insights - AI Insights provides an opportunity to identify the relationship between any given variables within your dataset. This will indicate a positive or negative correlative relationship, and provide a further detailed breakdown.




AI Hub - The homebase & tracker for any and all AI models that are used within any data flow. As the AI models are predictive in nature, we always advise being aware of where the historical data ends, and the predictive data begins.


Chat IVP - The Chat IVP function (not to be confused with the platform’s live chat) serves as a GPT enabled data scientist assistant that can support you in multiple ways, including the interpretation of your data. This is possible because the assistant has access to your dataset and can answer any broad or specific questions, bearing in mind the wider context in which the data sits.


Workspaces - Workspaces offers the ability to silo and compartmentalise different dataflows and dashboards. This will be the primary method of moving between multiple platform use cases such as sales data & finance data. Permissions can be changed in the gear cog in the top right corner of the screen, to permit or deny any users’ access to any given Workspace.


Having broken down the platform into its core components, it is worth familiarising ourselves with each feature & what they can do in more depth.