Advice and answers from the Advanced Web Ranking Team
>
>
Why ranking results may vary: AWR versus manual searches, similar tools, and Google Search Console
This article explains why ranking results from AWR can differ from manual searches, other ranking tools, and Google Search Console, focusing on factors like search engine definitions, geo-location, timing, and personalization.
When comparing the ranking results reported by AWR to those from manual searches, other ranking tools, or Google Search Console, differences may appear. These discrepancies are generally considered normal and usually arise from how each system retrieves and then processes the search engine results. Here’s an overview of why these differences may appear and why they are considered rather normal.
AWR versus manual search
The ranking positions in AWR are based on search results retrieved from SERPs when each update takes place, considering the search preferences selected at the search engine level (e.g., country-wide vs. custom location, language, and the types of results counted).
Search engine rankings are highly volatile, as algorithms have become more and more intelligent in determining user intent. Various factors influence result relevance, with location (the IP address from which the search is performed) and proximity to the searcher being among the most significant.
With each ranking update, AWR randomly selects the first available geo-located proxy IP in our system to query the chosen search engine. It then retrieves non-personalized SERPs from the closest data center to that proxy IP. The keyword queries sent by AWR match the exact words or phrases added to the project.
When performing a manual search in a browser, you are likely using your IP address or a VPN.
Since the IP location used by AWR differs from your IP, it is expected that the search results in your browser may not exactly match those reported by AWR.
Another key factor is timing. The moment you manually check rankings in a browser is different from the moment the rankings are updated in our system - depending on the updating frequency selected, the ranking system threads are created up to 24-48 hours before the rankings set is actually delivered. Once the queues are ready, the keyword queries are gradually sent until exhausted, in this mentioned range. Since rankings fluctuate throughout the day, a position checked manually at one moment may differ from AWR’s latest recorded position.
Other factors that can cause differences between AWR and manual search are:
Search engine definition: AWR computes the ranking positions based on the search engine definition you select. For example, if you choose Google Universal, all types of search results, including local packs, featured snippets, and PAA boxes, are counted in the ranking calculation, while if you select Google Organic, only featured snippets and organic results are included, excluding other result types like local packs. When comparing the ranking positions from AWR with the ones in the browser you should take into consideration which results are counted by the selected search engine in AWR.
Browser history and personalization: AWR collects non-personalized search queries, ensuring that ranking data is not influenced by prior searches or user-specific data. In contrast, manual searches may be affected by browser history, cookies, and personalization, which can skew the results
Browser add-ons: Add-ons or extensions in your browser, such as ranking trackers, can influence the way rankings are displayed and cause discrepancies between AWR results and manual search results.
AWR vs other ranking tools
There can be cases when the ranking positions displayed by AWR and other similar tools may vary, but this behavior is also considered normal.
Such discrepancies are generally due to the different proprietary querying mechanisms used by each tool. Each tool relies on its own search engine definitions, search parameters, proxies, and the actual timing of updates, which can all lead to reporting different ranking results.
For instance, while AWR may use a given set of geo-located proxy IPs to gather search data, another tool is likely to use a different set of proxies. Additionally, certain types of search results, such as local packs, featured snippets, or people also-ask boxes, may be included or excluded from reporting. Therefore, differences in search engine definitions can also lead to varying ranking results, even when querying the same search engine.
Since each tool uses its unique methodology for collecting ranking data, it’s normal to see some discrepancies between AWR and other tools.
AWR vs Google Search Console
It is also common to notice differences when comparing the ranking positions reported by AWR with those in Google Search Console.
AWR displays the ranking positions of the pages returned in SERP at the moment each update runs while GSC displays all the ranking pages where a keyword appeared in SERP throughout the entire day when the AWR update was completed.
Also, in GSC, the (average) position is calculated based on the best positions recorded by each keyword on each day across a selected date range.
For example:
If on day 1, a keyword query returns your website at positions 2, 4, and 6, its position is counted as 2 (the topmost position).
If on day 2, the same keyword returns your website at positions 3, 5, and 9, its position is counted as 3 (also the topmost position).
So, the average position across these two days is calculated from the topmost positions recorded each day: (2 + 3)/2 = 2.5.
In GSC, the positions reflect the best-ranking positions for the selected date (or range), whereas AWR captures the actual ranking for each update. While GSC is useful for analyzing keyword performance, it’s essential to understand that its reporting method is different from AWR’s. GSC uses an average of the best positions across a range of dates, whereas AWR focuses on exact, live data from the search engine on each update.
The ranking positions reported by AWR are always accurate, as they are based on live rank data retrieved directly from search engines using geo-located proxies. Since search results are constantly changing and influenced by multiple factors, it’s normal to see discrepancies when comparing AWR to manual searches, other ranking tools, or Google Search Console.
stay in the loop