Metamarkets

Increased performance leads to increased revenue. Our reporting feature lets buyers know ‘where things stand’ so you can make improvements to ‘get where you want to be’.

Oath Ad Platforms SSP takes pride in bringing you the best and most reliable technology. We deeply believe that great technology performance is critical to accelerating and growing your mobile advertising business.

Technology should deliver a strategic advantage and help you focus attention on business growth. But a baseline -of measured data points- is needed so you can move forward and make more money.

Guide decision-making with analytics. Our reporting features (via MetaMarkets) provide detailed insight and visibility to your transactions across our programmatic inventory.

  • Get detailed site, position, and impression-level attribute information about the premium inventory on the Oath Ad Platforms SSP.
  • Identify performance patterns across a variety of parameters and timeframes to help guide buying decisions.
  • Discover and analyze prior performance using simple, easy-to-understand visualizations.
  • Adapt to changing campaign needs easily with frequent data updates.

Start by selecting one of the tabs (explained below). There are several tabs available which are selected from the navigation bar.

Explore Tab

To different views are available under the Explore tab. The Nexage Auctions view is the default; to navigate to the My Bid Ops/Audience view, click the Explore button in the top left and choose My Bid Ops/Audience.

The Nexage Auctions view (default) populates upon initial access to Metamarkets. Use the Nexage Auctions dashboard to get a more in-depth information on the supply available on the SSP. Dig into specific publisher’s apps and sites and get pricing guidance per app bundle, compare supply sources at a high level from an auction, pricing, uniques, or CTR view, compare traffic by volume, such as iOS versus Android traffic, traffic breakdown by device country or more granularly at the DMA level, traffic by device type, traffic that does or doesn’t have device ID, does or does not have lat/long, and much more.

The My Bid Ops/Audience view is used to dig into your individual bidder’s performance across campaigns, creative IDs, publishers, seats, and more. It functions in the same manner as the Nexage Auctions view, but the data here is limited to only data for your bidder. Use this view to troubleshoot why you are losing an auction, whether it’s due to lost on price or bidding below the floor, to make bid adjustments, see campaign performance across different apps/sites, publishers, ad sizes, etc. to optimize targeting parameters, compare performance across creative IDs to optimize towards better performing supply, and more.

Each view under the Explore Tab has two areas; the time-series graph on the left and the dimension tables on the right. The data shown in the time-series graphs on the left is synced to the data shown in the dimension tables on the right. As you apply filters to one, the data shown in the other will automatically update to show the corresponding values. The dashboard is designed to shorten the time-to-insight by allowing you to perform high-speed queries across all data. Filters help shrink response time by narrowing the scope of queries to cover just the data of interest.

Time-series Graphs

Activity for the last 24 hours is displayed as a blue line on the charts. Customize the time period using the time-series filters. The default time-series provides a graphical representation of the last 24 hours of activity for each chosen metric.

Set Date Range

Change the time period used for analysis using the drop-down boxes located above the graphs. Beginning with default 24 hour view, zoom out or drill in using the thirteen options to see metrics over a longer period of time or to take a closer look at spikes and dips. Alternatively, click “Custom Range” and choose the date range from the calendar or by entering the dates to look at a specific date range not included in the available options. Drag the graphs to the left to expand the date range to previous days. Drag the graph to the right to expand the date range beyond selected days. Choose from hour or day with the right-most drop down above the time series graphs.

Compare Filter

Use the Compare view to drill into data by examining values across a range of metrics. To use the Compare filter, select the first date range and click the box next to “Compare” at the top of the graphs. Next, choose the date range to compare. All of the dimensions in the tables to the left will update with the percent deviation over the course of the date range.The graphs will also update to show the difference between the metrics over the date range. The blue line depicts the most recent date range and the gray line depicts the previous date range.

The Compare filter can also be used within a specific dimension. First click on a dimension box on the right side of the Explore page to expand it to show the top 250 values in that dimension ranked according to your chosen metric. Click the 'Compare' button located at the top of the dimension table to automatically select the top 4 dimension values for comparison on the time series charts on the left side of the page. Add or remove dimension values from your filter by clicking the values on the dimension tables (with a max of 7).

Exclude Filter

The Exclude filter provides additional options for slicing and dicing data. From the Explore page, click on one of the dimension tables on the right side. The box will expand to show the top 250 values for that dimension.

Select the Exclude button at the top of the box to enable a 'top-down' analysis of your dimension values. With the Exclude filter activated, click individual values to remove them from the analysis. Clicking the Exclude button a second time will re-select the value and include it in the analysis.

Remove the Exclude filter by clicking it again, or by clicking the X beside the value name in the filter summary at the top of the page. Upon switching back to the All Dimensions view or to another visualization page, the exclusions applied will stay in place unless removed.

Using the Filters Together

Use the time-series filter, to expand, shorten, or horizontally shift the time period used for an analysis. Once the time period is set, narrow the analysis by clicking individual values in the dimension tables on the right side of the page to filter across dimensions. Further dissect activity for a specific metric by clicking the header of a dimension box to zoom in to that dimension for a top-down or bottom-up analysis of the values within that dimension using the exclude and compare filters.

BACK TO TOP

Dimension Tables

The dimensions table, on the right, shows the top 12 entries for each dimension sorted by the chosen metric(s). If there are fewer than 12, all are shown.

To expand a dimension table, click on the + next to the table value name. To collapse a dimension table, click the - sign on the open table.

Filter the data across multiple dimensions by clicking the row in the tables. As additional filters are applied, they are summarized at the top of the page. Filters can be removed by clicking either the corresponding row or the filter's description in the summary (at the top of the page).

To further explore a specific dimension, click the name of that dimension at the top of its table. This expands the table to show all of the values for that dimension (along with the multiple metrics describing them), up to a maximum of 250.

Exploring a Dimension

In the expanded dimension's table, click the row(s) to narrow the view by dimension values. Select one or several. The time-series charts on the left will instantly update to show the aggregate time-series data for each metric.

Display a comparison view by clicking “Compare” at the top of the expanded dimension table once multiple dimension values have been selected. Dimension values will be plotted separately (instead of aggregating them), graph lines are color-coded for distinction. If “Compare” is clicked before specific dimension values are chosen, the top four are automatically displayed as determined by the chosen primary metric. Additional dimension values can be added to the comparison by clicking them. Up to seven dimensions may be compared simultaneously.

Selected values are implemented as a single filter that remains in place even if you navigate back to the “All Dimensions” view. Once filters have been applied you can save the view to your favorites, share it with a colleague, or download the time-series data.

BACK TO TOP

Using Filters in Other Views

Once a filter is applied, the keywords defining it appear at the top of the page. Unless removed, selected filters carry over to other views (such as pivot table), allowing filtered data to be viewed in other visualization formats.

Editing Filters

Filters can only be edited in the quick-filter menu. You cannot edit a filter directly; it must first be removed, then replaced by a new filter with the desired attributes. The quick-filter menu allows you to edit existing filters.

Removing Filters

There are several ways to remove a filter, 1) click the value used in the filter on the dimension table, 2) click the X beside the value name in the filter summary at the top of the page, or 3) click the Explore or Home tabs at the top of the page and reset the view.

Quick Filter +

Build tables, add and delete filters directly on the Explore page or using the Quick-Filter Menu located below the navigation bar and above the charts and tables sections. Click or unclick a dimension to filter by values.

There are two ways to create a filter: instant and batch. Create an instant filter by clicking the value name (not the checkbox), this immediately enables a one-value filter.

To create a batched filter with multiple values, select multiple values by clicking their checkboxes and clicking “Apply”. Batched values can be applied from only one dimension at a time.

BACK TO TOP

Sorting Tables by Most/Least Changed

Sort the values in a dimension table to show the top or bottom performers. To see the most or least changed entries in a period of time, sort the table by the actual or relative changes in value; this helps spot the biggest movers or least actives. Choose a comparison period from the date/time interval controls, to view relative (percent) changes (deltas) for all metrics in all dimension tables. Relative deltas appear only if a comparison period is set (no deltas appear if time periods are not being compared).

Deltas are calculated by comparing the current period's values to the corresponding values from the previous period. To view both absolute and relative deltas, and to sort the entries in a table by delta, first expand the table by clicking its title. By default, the entries in the newly opened table will not be sorted by delta.

Click the active relative or absolute delta control to change its sort direction, which is indicated by an arrow.

BACK TO TOP

Downloading Data From the Explore Tab

Explore views display historical metrics charts on the left and dimension tables on the right. You can download this data using the Download menu. The entries in the menu reflect whether all dimension tables or only when a single dimension is being viewed.

The menu options are:

  • Time Series Data – Downloads all data in the time frame and with the granularity specified in the time-series filters above the metrics charts. Only data for the metrics made visible in the metrics charts (by using the Metrics menu) is downloaded.
  • Time Series Data (all metrics) – Downloads all data in the time frame and with the granularity specified in the time-series filters above the metrics charts. Data for all metrics available in the view (regardless of whether or not they are displayed) is downloaded.
  • <dimension-name> Data – Downloads up to the top (or bottom) 250 rows for the dimension whose table is being viewed. The downloaded data is sorted in the same way as the dimension table. Rows contain only the metrics made visible in the dimension table (by using the Metrics menu).
  • <dimension-name> Data (all metrics) – Downloads up to the top (or bottom) 250 rows for the dimension whose table is being viewed. The downloaded data is sorted in the same way as the dimension table. Rows contain data for all metrics available in the view (regardless of whether or not they are displayed).
  • All <dimension> Data – Downloads all data available for the dimension whose table is being viewed. The downloaded data is sorted alphabetically by dimension name. Rows contain only the metrics made visible in the dimension table (by using the Metrics menu).
  • All <dimension> Data (all metrics) – Downloads all data available for the dimension whose table is being viewed. The downloaded data is sorted alphabetically by dimension name. Rows contain data for all metrics available in the view (regardless of whether or not they are displayed).

Troubleshooting Downloaded Data

Normally, dashboard data downloads work smoothly. If you do encounter an issue, it is likely fall into one of the following categories.

Unexpected Results in the Data File

If the download file does not contain the expected data set, check the following:

  • Is the range set by the time-series filters set to your desired range?
  • Did you choose the correct item from the download menu?
  • Does the dimension table you are downloading contain less than 100 entries?
  • If you did not wish to download all metrics, did you hide the unwanted metrics?

Note that viewing filters have no effect on downloaded data. For example, if you set a view filter to show only one publisher, then download the Publisher Name dimension, the downloaded data set will not be affected by that filter.

Slow or Failed Downloads

Not surprisingly, the size of the data set you are downloading can have a noticeable effect on how long the download takes. It is possible for a download operation to take a substantial amount of time for a data set encompassing hundreds of thousands of rows, for example. It's also possible for a browser to timeout while waiting to receive the CSV file, so that the download operation fails to complete.

To successfully download exceptionally large data sets, try the following:

  • Split the download into 2 or more parts can help. For example, if an attempt to download a week's worth of data is failing, consider downloading 2 to 3 days at a time.
  • If you are not interested in all of the metrics, use the Metrics menu to hide the unwanted ones and choose a download that does not include all metrics.
  • If you are mostly interested in top and bottom performers, consider downloading only the top and bottom 100 by sorting the target dimension by the desired metric, then downloading the top 100, resorting to flip the order, then downloading the bottom 100 (in each case not downloading all data).

BACK TO TOP

Facet Tab

Facet Table

Facet Table allows users to visualize up to 5 dimensions of their data in one table while optimizing for querying speed. This table also provides further interaction options, such as click filtering and collapsible dimension values.

To use Facet Table, choose Facet from the top drop-down menu and choose from the Auctions view or My Bid Ops view. If you choose the Auctions view, the data populated in your Facet table will be at the exchange level; adversely if you choose the My Bid Ops view, this data will be limited to data on your bidder.

Data Filtering

Filtering limits the scope of a selected data dimension to explore only the data needed. By selecting the “+” button next to “Filter” on the top left, you are able to add high level filters. For example, you want to only see data for one adomain and one creative ID, you can select these Filters so that the data displayed in the nested table below is only for these data points. You can dig into the data by adding dimensions.

Adding and Removing Dimensions in the Table

To add a dimension to the table, the user must click on the “Add Split” button in the top left corner of the table. To remove the dimensions displayed in Facet Table, click on the “x” sign in the dimension selector tabs at the top of the table.

Editing the Dimensions in the Table

In Facet Table, users can edit the dimensions of the table and its attributes related to the display of dimensions, including:

  • changing the dimension
  • sorting of dimension values: top or bottom
  • number of dimension values: 1 through 100
  • sort-by-metric: choose from the metrics displayed in the table

To remove a dimension, the user should click the “x” button that appears to left of the dimension value within the dimension box. To change the number of dimension values, the user should click the wheel button that appears to the right of the dimension value within the dimension box and choose from the values. To change the order of the dimension filters, the user can drag and drop dimensions to split as desired.

Add Metrics

You can add and remove metrics from the Facet Table by clicking on the “Add Metrics” box in the upper right hand corner and selecting or deselecting values from the drop down. These are the data points that will be displayed for the filters and dimensions you have in place. For example, if you wish to see data on “Bid Requests”, “Impressions”, and “Spend”, but do not wish to see data for “eCPM” and “Avg Win Bid”, you’d make these changes here.

Expanding and Collapsing Dimension Values

To visualize large data volumes with great speed, a nested table design is utilized. For example, let’s say a user creates a two-dimensional table. When the user adds the first dimension to the table, it looks like a simple single-dimension table. When the user adds a second dimension, the values of dimension 2 will be nested underneath the values of dimension 1. This nesting continues as more dimensions are added.

To view the values for dimension 2, the user can simply click on the plus sign to the left of the values for dimension 1 to expand the values underneath. A plus sign indicates that the dimension value can be expanded to show the values underneath. A minus sign indicates that the dimension value has already been expanded, but it can be collapsed to hide the values underneath.

Sort Chronologically by Time

In the Facet Table, users can not only sort the time dimension on a metric, but they can sort time chronologically.

To sort time chronologically, users can click on the time dimension tab to expose the dimension selector menu and change the sort-by metric to time. So, if a user has a table showing the time (hour) dimension, they can simply click on the time (hour) dimension tab to expose the drop down, and change the sort-by metric to time (hour).

BACK TO TOP

Download .csv Files on the Facet Tab

To download the data displayed in Facet Table as a .csv file, click the download icon in the top right hand corner of the page.

Note: there is a restriction on amount of data able to be downloaded. A rough way of calculating the amount of data that is allowed to be downloaded is to take the cardinality of all but the last dimension and then multiply them together; to be able to downloaded the cardinality needs to be less than 650. The cardinality allowance per dimension increases the fewer dimension splits you choose (ie. 500*500 works). The max you can download with 5 dimension splits is 5 values per dimension: 5 * 5 * 5 * 5 * X where X is any number.

Here are some examples:

2 dimension split of 500 > 500 = 500 queries < 650 (Under limit, can be downloaded)

3 dimension split of 50 > 50 > 5 = 2500 queries > 650 (Beyond Limit, cannot be downloaded)

3 dimension split of 5 > 100 > 500 = 500 queries < 650 (Under limit, can be downloaded)

BACK TO TOP

Heatmap Tab

A heatmap is a two-dimensional graphical representation that uses colors to denote the level of interaction between two variables. On the Dashboard, click the Views tab and select heatmap from the drop-down. The chart provides a quintile analysis of the interaction between two selected dimensions for a chosen metric. The darker the color of each box, the higher the value of that metric at the intersection of the two dimensions.

To download the underlying data for your heatmap, click "Download" in the top right hand corner of the page.

Toggle Clustering

Once the settings for a heatmap have been defined, press "Go" to populate it. By default, the heatmap will show the top values for each selected dimension in alphabetical order. To customize the analysis, press the "Toggle Clustering" button above the chart and reorder the values within each of the two dimensions (on the x and y axes) and organize them by the level of interaction between dimension values to best suit your needs. The dimensional values that are most similar to each other are grouped together or clustered to provide an analysis of dimensional values that act or behave similarly.

Note: filters that were applied on the Explore view (such as selection or exclusion) are carried over to the Heatmap view.

BACK TO TOP

Bar Chart Tab

Accessible through the Views tab, the Bar Chart View shows a columnar plot with values for a chosen dimension on the y-axis and a selected metric on the x-axis. Use the filters above the chart to quickly toggle between different dimensions and metrics, change the number of dimension values shown on the y-axis, or to switch the x-axis scale from standard to logarithmic.

To download the underlying data for your bar chart, click on "Download" on the top right hand corner of the page.

BACK TO TOP

Bookmarks

Bookmark and save any view or analysis within the application and quickly return to it at a later time. To save a view as a “bookmark”, click the bookmark icon in the top right corner. Or bookmark the page using the browser's 'Bookmark' or 'Favorites' feature. Filters and settings are retained.

Bookmarks can be renamed or deleted in saved view. Click the solid bookmark icon while in bookmarked view to see the options for deleting or renaming the bookmark.

All dashboard views can be downloaded as a CSV file containing the underlying data. The historical data in this file can then be further processed and analyzed, for example by being loaded into a spreadsheet.

Access a list of bookmarks by clicking the “Bookmarks" dropdown in the banner at the top of your Dashboard.

BACK TO TOP

There are two ways to search your Dashboard, clicking a search result generated by any of these methods creates a filtered view on the Explore page.

  1. Click the magnifying glass located at the top right-hand corner of the Dashboard, to open the search box. Search one dimension or all dimensions for a given attribute, such as a keyword or number. Enter a keyword and click Search to generate a list of possible matches organized by dimension.
  2. Expand the Dimension on the Explore page to expose a search option at the top right of the dimension table. Click it to open a search box which can be used to search the entire table (not just the 250 displayed entries). Click a search result to view it in the Dashboard.

BACK TO TOP

For additional How To information on Metamarkets, visit this support link for short videos on how to best navigate and use the Metamarkets dashboard.

 

Have more questions? Submit a request