SKU:
DPK-POIADD-BKLThis Add On leverages proprietary algorithms to provide comprehensive contextual analysis by comparing and benchmarking POIs within a 5 km radius, up to a maximum of 25 similar POIs.
Our proprietary algorithm identifies similar POIs based on industry and distinctive characteristics, calculating average scores for key performance indicators (KPIs) such as sentiment, popularity, main clusters, and discussed topics over a 12-month period.
The Benchmark Add-On enhances POI contextual understanding and supports data-driven decision-making and strategic planning. Stay tuned for further updates as we continue to expand our capabilities.
Field Name | Defintion | Type | Example |
poi_id 🔑 | Unique identifier of a POI. Use it to recognise the point of interest or as a foreign key to link different files and/or packages together. | STRING | 0056912b351fd6b97979101a7d3a2004c31dceb7 |
benchmark_sentiment | Proprietary percentage KPI (valid range from 1 to 100) which measures the level of customers’ global satisfaction for POIs similar to the one being analyzed within a 5 km radius. The score is calculated over a period of 12 months, starting from the value of the date_refreshed attribute, and is obtained through a proprietary semantic algorithm that considers multiple attributes of the online content (feedback, topics, opinions, languages…). | DOUBLE | 76.7 |
benchmark_popularity | Proprietary percentage KPI (uncapped range) that measures the popularity of POIs similar to the one being analysed within a 5 km radius. The score is calculated over a period of 12 months, starting from the value of the date_refreshed attribute, and is obtained through the analysis of multiple factors such as the number of geo-localized reviews and social media content. This KPI can be used for comparison of POIs belonging to the same industry. For more info on Industries and Categories, please see this page.
Instead, for info about the Popularity Index click here. | DOUBLE | 90.45 |
benchmark_main_clusters | Array in which each element represents a binary pair composed of Benchmark Sentiment KPI (valid range from 1 to 100) and cluster. Each value in the array is calculated over statistically reliable clusters, a super set of topics on which a user expresses opinions, related to POIs similar to the one being analyzed within a 5 km radius. These clusters can be inherent to multiple industries and/or categories. All elements are sorted in alphabetical order based on the name of the cluster, and the array can contain multiple elements that correspond to the most relevant clusters identified. For each cluster, the score is calculated over a period of 12 months, starting from the value of the date_refreshed attribute, and is obtained through a proprietary semantic algorithm that considers multiple attributes of the online content (feedback, topics, opinions, languages…). Please note that a cluster name is supplied in English by default; it is possible to request another language among the following: French, German, Italian, Spanish, Portuguese. | ARRAY (JSON) | [{"cluster": "Appearance","sentiment": 74.29},{"cluster": "Atmosphere","sentiment": 64.05},{"cluster": "Cleanliness","sentiment": 33.33},{"cluster": "Costs","sentiment": 6.67},{"cluster": "Food Quality","sentiment": 81.67},{"cluster": "Hospitality","sentiment": 77.66},{"cluster": "Position","sentiment": 94.74}] |
benchmark_most_discussed_topics | ### Definition of Benchmark Topics
JSON representation of the top 5 most discussed topics related to POIs similar to the one being analyzed within a 5 km radius. Each element represents a binary pair composed of Benchmark Sentiment KPI (valid range from 1 to 100) and topic. Each value in the array is calculated over statistically reliable topics, a subject on which a user expresses opinions, that can be inherent to multiple industries and/or categories. All elements are sorted in ascending order based on the value of the Sentiment KPI, and the array can contain a maximum of five elements that correspond to the most relevant topics identified. For each topic, the score is calculated over a period of 12 months, starting from the value of the date_refreshed attribute, and is obtained through a proprietary semantic algorithm that considers multiple attributes of the online content (feedback, topics, opinions, languages...). Please note that a topic name is supplied in English by default; it is possible to request another language among the following: French, German, Italian, Spanish, Portuguese. | ARRAY (JSON) | [{"topic": "food","sentiment": 76.19},{"topic": "service","sentiment": 80.95},{"topic": "restaurant","sentiment": 68.42},{"topic": "local","sentiment": 66.67},{"topic": "personal","sentiment": 86.67}] |