Top Concepts - v1

Strategy properties

Strategy id top_concepts
Version 1
Supports grouping? Yes
Learn more...
Cost (#groups x 6)
Learn more...

Strategy usage

Overview

This strategy identifies the concepts mentioned the most in content that is seeing engagement from an audience. The idea of concepts is explained in detail in the Working with Concepts guide for the data source.

The strategy allows you to select an audience using demographic attributes, by a topic being engaged with, or both. The strategy also allows you to specify an audience to compare your audience against.

By comparing the concepts for your audience against a comparison audience you can see which concepts are uniquely engaging for the audience, as opposed to concepts that are popular in general across LinkedIn.

For each concept identified the strategy tells you:

  • the name and type of the concept.
  • the number of members who have engaged with content relating to the concept.
  • relatively how engaged your audience is with the concept compared to your selected comparison audience.
  • the relevance of the concept for your audience (based upon the number of members who have engaged with the concept, and relatively how engaged your audience is with the concept compared to the comparison audience).

Use cases

The strategy helps you answer questions such as:

  • For a defined audience, which concepts are mentioned the most in content that is driving engagement?
  • For a defined topic, which concepts are mentioned the most in content that is driving engagement?

Parameters

To execute this strategy without grouping you must provide specify at least one of the keywords, concepts, or audience parameters. To execute this strategy with grouping you must specify the groups parameter.

Parameter Type Required? Description
audience audience

At least one of 'keywords', 'concepts', and 'audience' must be given.

Selects the audience to analyze.

concepts concepts

At least one of 'keywords', 'concepts', and 'audience' must be given.

Concepts for the topic to analyze.

keywords keywords

At least one of 'keywords', 'concepts', and 'audience' must be given.

Keywords for the topic to analyze.

exclude_types

array(string)

No. Defaults to ["politician", "entertainer", "sportsperson", "person"].

Excludes the specified types of concepts from the analysis result. By default types that represent people are excluded.

include_types

array(string)

No.

If supplied this parameter overrides exclude_types, and only concepts of the types specified will be included in the result.

period period

No. Defaults to 28 days.

The time period to analyze.

sort_by sort_by

No. Defaults to relevance.

Note that for this strategy "relevance" is an additional sort option.

The output field to sort results by.

comparison_audience comparison_audience

Yes.

The audience to compare engagement with.

groups

No.

Groups to perform analysis for.

See Learn more...

Example requests

The following examples show which parameters should be submitted to the API to answer the stated questions.

Which concepts are mentioned the most in content seeing engagement from members in the US for the topic of 'cloud', compared to the global audience?

{
    "type": "strategy",
    "name": "example",
    "subscription_id": "cd99abbc812f646c77bfd8ddf767a134f0b91e84",
    "parameters": {
        "strategy": "top_concepts",
        "version": 1,
        "parameters": {
            "keywords": {
                "any": ["cloud", "azure", "aws", "google cloud", "vmware"]
            },
            "audience": {
                "countries": ["united states"]
            },
            "comparison_audience": "global"
        }
    }
}

Which concepts are mentioned the most in content seeing engagement from decision makers in the UK for the topic of 'cloud', compared to the general UK audience?

{
    "type": "strategy",
    "name": "example",
    "subscription_id": "cd99abbc812f646c77bfd8ddf767a134f0b91e84",
    "parameters": {
        "strategy": "top_concepts",
        "version": 1,
        "parameters": {
            "keywords": {
                "any": ["cloud", "azure", "aws", "google cloud", "vmware"]
            },
            "audience": {
                "countries": ["united kingdom"],
                "seniorities": ["manager", "director", "vp", "cxo", "partner", "owner"]

            },
            "comparison_audience": {
                "country": "united kingdom"
            }
        }
    }
}

Here the parameters request that only members in the UK in senior positions are analyzed and their engagement is compared to all members in the UK.

Which companies and products are mentioned the most in content seeing engagement from members in the US for the topic of 'cloud', compared to the global audience?

{
    "type": "strategy",
    "name": "example",
    "subscription_id": "cd99abbc812f646c77bfd8ddf767a134f0b91e84",
    "parameters": {
        "strategy": "top_concepts",
        "version": 1,
        "parameters": {
            "include_types": ["product", "company"],
            "keywords": {
                "any": ["cloud", "azure", "aws", "google cloud", "vmware"]
            },
            "audience": {
                "countries": ["united states"]
            },
            "comparison_audience": "global"
        }
    }
}

Here the 'include_only' parameter filters results to only concepts of type 'product' or 'company'.

Output

If the strategy is successfully executed it will return output with the following structure:

{
    "redacted": false,
    "unique_authors": 92000,
    "interactions": 103000,   
    "concepts": [
        {
            "concept_type": "company",
            "concept_name": "Microsoft",
            "unique_authors": 10000,
            "interactions": 12000,
            "engagement_ratio": 1.4,
            "relevance": 68.01
        },
        {
            "concept_type": "product",
            "concept_name": "Skype",
            "unique_authors": 13000,
            "interactions": 11300,
            "engagement_ratio": 1.23,
            "relevance": 57.22
        }
    ]
}

If the redacted property is true then not enough data was found for the analysis. Consider widening your topic or audience for your analysis.

The following properties are returned for each concept identified:

  • concept_type - the type of the concept identified.
  • concept_name - the name of the concept identified.
  • unique_authors - how many members engaged in with the concept in the analyzed period.
  • interactions - how many engagements related to the concept in the analyzed period.
  • engagement_ratio - relatively how engaged your audience is with the concept compared with your selected comparison audience.
  • relevance - the relevance of the concept for your audience (based upon the number of members who have engaged with the concept, and relatively how engaged your audience is with the concept compared to the comparison audience).

Grouping

The strategy supports analysis by multiple groups. Learn how to run strategies with grouping by reading our Performing grouped analysis developer guide.

As an example the following parameters will analyze the top concepts for the topic of 'cloud' for each of the two countries specified:

{
    "type": "strategy",
    "name": "example",
    "subscription_id": "cd99abbc812f646c77bfd8ddf767a134f0b91e84",
    "parameters": {
        "strategy": "top_concepts",
        "version": 1,
        "parameters": {
            "keywords": {
                "any": ["cloud", "azure", "aws", "google cloud", "vmware"]
            },
            "comparison_audience": "global",
            "groups": {
                "list": {
                    "countries": ["united kingdom", "france"]
                }
            }
        }
    }
}

When run with grouping the output of the strategy gives a set of results for each group within the 'groups' property. For example:

{
    "groups": {
        "united kingdom": {
            "redacted": false,
            "unique_authors": 92000,
            "interactions": 103000,
            "concepts": [...]
        },
        "france": {
            ...
        },
        ...
    }
}

Usage costs

The cost of running the strategy is based upon the number of groups you specify.

If you run the strategy without grouping then the cost is 6 analysis tasks.

If you run the strategy with grouping then the cost is:

Analysis tasks consumed = 6 x (number of groups)

note icon

Note that if you specify groups for your analysis using the 'top' parameter the strategy will run an extra analysis task to fetch the top segments to be used for the grouped analysis. As a result one additional analysis task will be consumed.