Relevance ratings can be given starting from Navigational (if applicable) to Bad.
&
modifier? We can use official resources like a businesses’ official websites, national postal service
websites, and government websites to confirm the user intent.
doing this we would need to use the information we have gathered from the first step and use real world research to determine the relevance of the result that is returned. >
If the user’s location intent is not expressly stated in the query, for example [chinese] or [479 margarita ave], use the user location, viewport location, and viewport age to determine the area of expected results.
Viewport | User | Location Intent |
Fresh | Inside Viewport | When the user is within a fresh viewport, take the user location as location intent. Re:suits are generally expected in or near the viewport, anc re:sultsinsic ie the area cannot be rated Bad because of distanee alone. |
Outside Viewport | Results are expected in or near the viewport area. All relevant results inside the viewport are eligible for;a rating of Excellent. If no results can be found in or near the viewport, consider the user location a secondary location intent. | |
Missing | When the user is missing, the viewport is the location intent. | |
Stale | Inside Viewport | When the viewport is stale, consider only the user location as location intent. |
Outside Viewport | ||
Missing | Use the stale viewport as location intentwl hen the user location is missing. | |
Viewport Age Missing | Present or Missing | Consider the viewport fresh when thle viewport age is missing. |
Missing entirely | Present | The user location set:s the location intent when the viewport is not present. ^> |
Missing | When the user location and viewport are missing, the test locale becomes the location intent with a strong focus on prominent results. |
Previously the guidelines contained the concept of ‘Associated Entities’. The concept of Associated Entities has been removed and no longer applies. This means that a result cannot satisfy user intent only because of its containment relationship with the POI/business in the query. Therefore, according to the guideline section 5.1.8. Lack of Connection, If the result does not satisfy the user intent either because there is no relationship between the query and the result or because the connection between the query and the result will not be immediately obvious to the user, rate the relevance Bad. The relevancy rating for this result is therefore Bad.
User Query | Result | Type of Connection | Description |
[airport] | Santa Cruz Boardwalk Santa Cruz | None | Does not satisfy user intent: Research shows that there5 is no connection between the query andtteresUt. Rate Bad. |
[Raging Waters 2333 South White Rd San Jose] | 2333 South White Rd San Jose | None | Does not satisfy user i ntent: Result is for the correct address, 1:31jt does not include the business named in the query, so users will have no way of know ing if this is truly connected to the business or not. Rate Bad. |
[Raging Waters] | 2333 South White Rd San Jose | None | Does not satisfy user intent: Query is the name of a business and result is for the correct address, but does not include the business name. Users will have no way of knowing whether or not this address is connected to the business they asked for. Rate Bad. |
[valley fair mall] | Macy’s 2801 Stevens Creek Blvd, Santa Clara, | None | Does not satisfy user initenit: Result is for a store, not a mall. This store is insidethe requested maII, but users will 1nave no way of knowing this simply fre>m looking at the result. Rate Bad. |
[macy’s] | Westfield Valley Fair 2855 Stevens Creek Blvd, Santa Clara, | None | Docs not satisfy user intent: Result is for a mall, not a store. This mall contains the requested store, but users will have no way of knowing this. Rate Bad. |
Example 1 – 5.16.1 Transit Queries
Location and Real-World Results
Our result is a train station in the neighboring locality Ravenstein. Section 5.16.1 of the guidelines explains how to rate relevance for this result.
This task is similar to the Stockport example. Other stations that could satisfy intent should be demoted for distance. There is a train station that is called Oss that would fully satisfy intent. Our result is a train station near Oss, which could still be relevant. Relevance is demoted -1 for distance to Good. https://bit.ly/3jMpBve
Example 2 – 10.7 Category Queries
Viewport Center:
52.282972, 4.812925
User Location: 52.272949, 4.812012
Here the user is looking for [Bouwmarkten]. This is a category query without a location modifier. Section 2.3.2 of the guidelines explains that when the location is not stated in the query, the user location, viewport and viewport age determine the location intent. In this case, the user is outside of the fresh viewport, so the viewport is the location intent.
Location and Real-World Results
Next, we can take a closer look at the Kwantum website to see how well the result fits the category.
Now we must look for possible results near the viewport that can affect our relevance rating. If we look for hardware stores near the viewport location in a map application, we can see that there are a few DIY/hardware stores near the viewport. Our result, the Kwantum store, is one of the closest results, so we do not have to demote relevance for distance.
Kwantum does offer some products and services that one would expect from a hardware store, but their products for this category are quite limited compared to other chains such as Praxis or GAMMA. Section 10.7 explains that results that technically fit the category, but do so less well than other possible results, should be demoted. In this case, relevance should be demoted -1 or -2 for a user intent issue to Good/Acceptable.
Viewport Center:
51.641749, 4.829019
User Location:
51.657715, 4.812012
In this task the user is looking for [Tuincentrum]. This is once again a category query without a location modifier. The user is outside of the fresh viewport, so we are looking for garden centers near the fresh viewport.
If we research the result business by going to their official website, we can see that the result is a tree nursery that mainly sells to other businesses. Regular customers are appointment only.
Location and Real-World Results
https://www.boomkwekerijakkermans.nl/
Section 10.7.3 explains that when the query is for a clear and straightforward category, results must belong to that category. Our result is a tree nursery, which does not fit the category of garden centers. Because our result is a tree nursery and not a garden center, Relevance is demoted to Bad due to user intent.
Voor groteenkle.net. inen
Wij late vorm bomen zelf in de juiste vorm groeien. We gebruiken hiervoor Sterke bamboerekken
waarwedel:akken in de gewenste vorm aan vastbinden. Naast standaardmaten, leveren we ook maatwerk: u kunt bij ons aangeven hoe groot de boom mag zijn. Precies passend in uw tuin.
Wanneer bomen planten?
De beste tijd om een boom te planten isvan half Oktober tot half april.We kunnen de bomen tot in uw tuin bezorgen, zodat u ze zelf makkelijk kunt planten. U krijgt van ons plantadvies, verzorgingstips en snoei-instructies mee. Hiermee gaat het planten zeker lukken.
Kies uw eigen boom
Zoekt u een grote boom.of een kleine? Wilt u extra schaduw in de tuin of meer privacy? Een boom met een bijzondere vorm? U bent op afspraak van harte welkom op onze kwekerij in Oosterhout.
Bel 0162 45 32 04 of mail naar [email protected].
Example 4 – 10.3 Query Address Does Not Exist
Viewport Center: 52.689180, 5.749370
User Location:
52.646484, 5.800781
In this task the user is looking for a full address. Section 10.1 Specific Address explains how to rate results for a query like this. The user is looking for a specific location, so the user location and viewport are irrelevant. A result for this exact location should be rated Navigational. If we look at our result, we can see that the result is for the same street, but a different street number. The query is for number 21L, while the result is for number 21J. Before we can rate relevance, we should research the query and result.
Location and Real-World Research
A reliable resource for checking addresses is https://postcode.nl. If we use this site to look up Verlengde Gildenweg 21L, it appears that this street doesn’t have number 21L.
https://www.postcode.nl/8304BK/3-27
Section 10.3 explains how to rate results when the query is for an address that doesn’t exist. If the query address doesn’t exist and the result is one of the closest possible verified addresses, relevance should be rated Excellent.
Example 5 – 10.3 Query Address Does Not Exist
Viewport Center: 55.484682, 23.896703
User Location: 52.272949, 5.603027
This is a query for a partial address. The query only includes a street name and number, but no locality.
Location and Real-World Research
To research the query, we can use https://postcode.nl again. It appears that there is no street called Litouwensestraat. Only a street called Litouwenstraat exists. If we look at this street, we can see that this street does not have number 23. It only has number 22. The query likely is misspelled for Litouwenstraat 22, because there are no other possible matching addresses.
https://www.postcode.nl/straat/295747/Litouwenstraat
The result is an address result for this same street, but it does not include a street number. The result is just for the street, while the query is for a specific street number. Section 10.3 also explains how to rate relevance in this specific case. When the result is just the queried street without a street number, relevance is rated Acceptable.
Example 6 – 10.2 Non-Specific Address
Viewport Center: 51.493046, 5.396293
User Location: 51.435242, 5.476685
In this task the user is looking for a street called Julianastraat without a locality. Section 10.2 Non-Specific Address explains how to rate results for
a query like this. When the query is a partial address, the user is looking for locations that are closest to their location or the
viewport.
Location and Real-World Research
The user is outside of the fresh viewport. Section 2.3.2 of the guidelines explains that in this case the viewport is the location intent. We will have to look for streets called Julianastraat near the viewport before we can rate relevance for this result.
If we go to the viewport location in a maps application and begin looking for Julianastraat, we can already see some streets with the same name in Eindhoven and Sint-Oedenrode that are closer to the viewport than our result in Diessen. Because there are a few streets that are closer to the viewport, relevance for this result is demoted for distance to Good or Acceptable.
Example 7 – 10.6.2 Chain Businesses
Viewport Center:
51.497721, 3.610066
User Location: 51.495667, 3.609009
In this task the user is looking for H&M stores. The query does not include an explicit location and the user is inside of the fresh viewport, so the user location is the location intent. We’re looking for H&M stores near the user location.
Location and Real-World Research
Our result is an H&M store that appears to be the closest result, however, as sections 1.3.3 and 10.6.2 of the guidelines explain, we should always rate against the real world. If a better result exists that isn’t shown in Tryrating, we should demote the result accordingly. Because of this, we have to find all H&M stores near the user before we can begin rating relevance.
To find all stores near the user, we can use the official H&M website. If we look for stores near the user location in Middelburg, we can see that there is in fact a store in Middelburg that isn’t shown in Tryrating. Our result is the second closest result, so relevance should be demoted -1 for distance to Good.
Example 8 – 5.14 Unexpected Results
Viewport Center: 52.095351, 4.394897
User Location: 52.097168, 4.416504
In this task the user has entered a straightforward query for the locality Nieuwerkerk aan den IJssel. The user intent is clear.
The result is the main train station in Nieuwerkerk aan den IJssel. This might not be what the user is looking for, but it can still satisfy a secondary intent. Section 5.14 Unexpected Results explains how to rate relevance for this POI. Because this train station has the same name as the locality and could satisfy a secondary transit intent, relevance is demoted to Good for user intent.
Example 9 – 5.14 Unexpected Results
Viewport Center: 52.054149, 4.492801
User Location: 52.044348, 4.517897
Example 9 – 5.14 Unexpected Results
In this task the user is looking for Almere Buiten. It might not be immediately obvious what the user is looking for, so to be sure we can do some research.
Location and Real-World Research
https://nl.wikipedia.org/wiki/Almere_Buiten
Almere Buiten is a district of the city Almere. The most likely user intent is this district. The result is one of the train stations in Almere Buiten and it has the same name. This task is similar to the previous task. The primary intent is the district, but this train station could satisfy a secondary transit intent (guidelines 5.14 Unexpected Results). Relevance is demoted to Good for user intent.
Example 10 – 10.6.3 Chain Business with Location Modifier
Viewport Center: 52.353632, 5.630430
User Location: 52.492676, 5.515137
Here the user is looking for McDonald’s in Zoetermeer. This is a query with a location modifier.
Section 2.3.1 explains that when the query includes a location, the user location and viewport should be ignored. For this query, we’re looking for McDonald’s restaurants near Zoetermeer. Section 10.6.3.1 General Location Modifier explains how to rate results for this query. Relevance is not demoted for distance for results inside of the specified location. Results outside of the location will be demoted based on how many results there are
Location and Real-World Research
Before we can begin rating relevance, we should do some research and look for all McDonald’s restaurants in the real world. We can use the official McDonald’s website to find all restaurants near Zoetermeer.
On The McDonald’s website we can see that there are a few restaurants inside Zoetermeer. Our result is located just outside of Zoetermeer in Bleiswijk. Because of this, relevance is remoted -1 for distance to Good.
Viewport Center: 51.700942, 7.309305
User Location: 51.440735, 5.509644
In this task the user is looking for a street name and number. The query does not include a specific location. Section 10.2 Non-Specific Address explains that in this case the location intent should be based on the user location and viewport.
Location and Real-World Research
The user is outside of the fresh viewport, but the viewport is located in Germany, where we won’t find any Dutch addresses like the requested address. Section 2.3.2 of the guidelines explains that when there are no results near the viewport, the user location is a secondary location intent. Next, we will look for possible locations for this query. If we use http://postcode.nl to look for this address, we can see that there is only one street called Lombokpad. If we look further, we can see that number 238 does not exist.
The query might be for a non-existing address, or it might be misspelled for Lombokpad 2-38, which does exist. Either way, the result is for the street without a street number.
Section 5.1.7 Address-Result Connection explains that when the query is for a full address and the result is only the street, relevance should be rated Acceptable.
Eindhoven, Lombokpad
Is dit niet wat u zoekt? Bekijk alle resultaten.
Postcode | Straat | Reeks | Woonplaats | Gemeente |
5641BZ | Lombokpad | 1-3 | Eindhoven | Eindhoven |
5641BZ | Lombokpad | 2-4 | Eindhoven | Eindhoven |