|
This article covers the following: |
Overview
When comparing VWO reports with those from third-party analytics tools such as Google Analytics, HubSpot, Mixpanel, or an internal CRM/database, you may notice differences in visitor counts and conversion rates. This discrepancy is a common occurrence and rarely indicates an error in either tool. Instead, these discrepancies usually arise from fundamental differences in how each platform is configured and how it tracks and processes data.
This article provides a comprehensive checklist to help you identify potential reasons for these differences, enabling you to align your data more effectively and draw more accurate conclusions.
Potential Causes for Data Discrepancy
This section outlines the most likely causes for the data differences you are observing.
Note:
- The causes listed below are not in a specific order of importance or likelihood. A data discrepancy can result from a single factor or, more often, from a combination of several factors.
- A 5–10% difference is generally expected when comparing VWO data with third-party analytics tools. This is because different tools use different methods to track visitors, define sessions, filter bots, and process data. As a result, minor variations in visitor counts and conversions are normal. If you see a larger discrepancy, such as 20–30% or more, use the checklist below to investigate potential causes.
Issues with VWO SmartCode Setup
You can often trace discrepancies back to the initial setup of the VWO SmartCode on your website. The table below outlines common SmartCode-related issues that can affect tracking and reporting.
| Potential Cause | Impact | Recommended Action |
| Missing VWO SmartCode | If the SmartCode is missing on all pages tracked by your third-party tool, VWO cannot record visitors or conversions on those pages, resulting in lower numbers. | Ensure that the SmartCode is installed correctly and consistently across all pages involved in your campaign, including conversion pages. |
| Inconsistent Setup | Security protocols like Content Security Policy (CSP) can block VWO’s script from executing, leading to tracking failures on certain pages or for specific users. |
Review your website's security settings and browser console logs for any errors that might be blocking the VWO script. Ensure that you have whitelisted VWO's domains if necessary. For more information, see How to Whitelist VWO in Your Website’s Content Security Policy (CSP)? |
Inconsistent or Incorrect Visitor Identification
VWO and third-party tools use different logic to identify and count unique visitors, which can result in the same user being treated differently across platforms. The following table highlights visitor identification scenarios that commonly lead to data differences.
| Potential Cause | Impact | Recommended Action |
| Visitor vs. Session Tracking | VWO uses a persistent cookie (UUID) to identify unique visitors, treating an individual as a single visitor across multiple sessions. Many analytics tools are session-based, counting the same user as a new visitor if they return after a specific period. This leads to higher visitor counts in session-based tools. | Use VWO for campaign-centric conversion rates (which are visitor-based). When comparing data with a third-party analytics tool, look at unique users/unique visitors instead of sessions to ensure you’re comparing similar metrics. |
| Cross-Device Identification | Some tools attempt to identify the same user across different devices, such as phone and desktop, whereas VWO's primary identification is per-device’s browser. This difference in configuration can lead to different unique visitor totals. | Understand the visitor identification model of each tool. Comparing a device-centric count (VWO) with a user-centric count (from some third-party tools) will always reveal a discrepancy. |
Misconfigured Campaign or Page Setup
Campaign eligibility rules, such as targeting conditions, page triggers, and mutual exclusivity, play a key role in determining who views a campaign. The table below lists configuration-related factors that can influence campaign data.
| Potential Cause | Impact | Recommended Action |
| URL Targeting Rules | Your VWO campaign may run on a particular set of pages, for example, example.com/product/a, while your analytics tool tracks a broader pattern, for example, all pages under /product/. This will naturally result in different visitor counts. |
Ensure that the URL targeting and exclusion rules in your VWO campaign precisely match the pages or segments you are analyzing in your third-party tool. Check query parameters and ‘/’ inclusion/exclusion settings. For more information, see Targeting and Segmentation in VWO. |
| Split URL Test Configuration | In a split URL test, some third-party tools might count a visitor on the control page and again on the variation page, resulting in a double count for the same visitor. |
This is a known difference in tracking logic. Acknowledge that the third-party tool may inflate visitor counts for split URL tests. This behavior occurs because the visitor first loads the control page and is then redirected to the variation page. As a result, some third-party analytics tools may register two page loads for the same visitor, which can inflate visitor counts for split URL tests. To minimize this discrepancy, consider configuring your analytics tracking so that the analytics script executes after VWO determines the visitor’s variation. This can be achieved by using custom code to ensure VWO runs before the analytics tool on the page. For more information, see How to Implement Event-based Redirection for Split URL Tests? |
| When reviewing page-level data in a third-party analytics tool, visitors who are redirected during a split URL test may appear as separate page loads. This can inflate visitor counts and make direct comparisons with VWO reports difficult. |
Consider integrating the two tools and reviewing the VWO-pushed campaign data. When the integration is enabled, VWO tags visitors who participate in a campaign and sends this information to the corresponding third-party tool. You can then create reports in that tool to review campaign-level and variation-level data. For more information on setting up integrations between VWO and third-party analytics tools, see VWO Anlaytics Integrations. |
|
| Traffic Allocation | If your VWO campaign is set to run on less than 100% of your total traffic, VWO will only report on that portion of visitors. Your third-party tool, however, will report on 100% of the traffic, creating a predictable discrepancy. | When comparing, be aware of the traffic allocation percentage in VWO. Some tools may not have a feature to filter by the same percentage, so a direct comparison will be skewed. |
| Mutually Exclusive Groups (MEG) | If your campaign belongs to a Mutually Exclusive Group, visitors who have already been exposed to another campaign in the group are excluded from this campaign. As a result, VWO reports fewer visitors, while the third-party tool continues to track those visitors on the page. | When using MEG, expect lower visitor counts in VWO campaigns compared to a third-party tool that does not use this exclusion logic. |
| Campaign Activation Triggers | Campaigns can be triggered based on specific conditions, for example, time spent on a page or element click. If a visitor leaves the page before the trigger condition is met, VWO will not count them. An analytics tool, however, may have already counted them as a page visitor. This also applies to Manual Campaign Activation in VWO. | Ensure you consider the trigger logic. A third-party tool might track all page visits, while VWO only tracks visitors who meet the specific activation criteria. |
Audience Segmentation or Environmental-Related Mismatches
Visitor attributes and environmental signals may be interpreted differently by each analytics tool, affecting who is included in reports. The table below covers audience- and environment-related factors that commonly contribute to discrepancies.
| Potential Cause | Impact | Recommended Action |
| Target Audience Segmentation | A VWO campaign targets a specific audience, such as visitors from New York or users on mobile devices. When a third-party tool does not apply the same audience filters, it reports higher numbers. | To compare accurately, apply the same segmentation filters (device, location, user behavior) in your third-party tool that you have used for targeting in your VWO campaign. |
| Device Detection Logic | VWO and other tools use different libraries and methods to detect a user's device type (desktop, mobile, tablet). One tool might classify a large tablet as a desktop, while another might classify it as a tablet, leading to mismatched device-specific reports. | This is an operational difference between tools. Be aware that device-specific segments may not align. |
| Traffic Source Detection Logic | Different tools may use different logic to identify and categorize traffic sources. For example, one tool may rely solely on UTM query parameters to classify paid traffic, while another may also factor in referrer data or other signals. As a result, tracking numbers can vary across traffic sources when comparing data from different tools. |
This is an operational difference between tools. As a result, traffic source segments may not always align across platforms. For more information on how VWO categorizes traffic sources, see Traffic Attribution Hierarchy. |
| Geolocation Identification | VWO uses a specific IP address detection service (Maxmind) to determine a visitor's location. Other tools may use different services or methods, which can result in a visitor being assigned to a different city or region, especially if they are near a border. | Minor differences in location-based segments are expected due to the varying accuracy of IP lookup services. |
Delays or Gaps in Data Processing
After data is collected, each tool applies its own processing, filtering, and validation logic before finalizing reports. The table below summarizes data processing–related reasons that can lead to differences in reported numbers.
| Potential Cause | Impact | Recommended Action |
| Bot Filtering | VWO has a built-in bot filtering feature to exclude non-human traffic. If your third-party tool uses a different bot-filtering mechanism (or none at all), it may report higher traffic numbers by including bots that VWO has excluded. | Understand that VWO provides cleaner, human-only data, which may be lower than unfiltered data from other tools. |
| Campaign/Account Timeline Changes | If you make significant changes to a live campaign, for example, altering targeting rules or changing traffic allocation, VWO may flush the existing data to ensure the new report is accurate. This data flush resets the numbers in VWO, but your third-party tool's data remains cumulative. |
Be aware that making critical changes to a live campaign can reset its data in VWO.
When comparing, align the date range in your third-party tool to start from the time the campaign changes were made. |
| VWO SmartCode Timeout | VWO uses a timeout to prevent its script from slowing down your website. If a visitor's network is very slow and the script does not execute within this timeout period, VWO will not track them, but your third-party tool might. |
This is a necessary trade-off for performance and affects a very small percentage of users.
Recognize that certain performance optimizations affect a small percentage of users and can lead to minor, expected discrepancies. |
| Blocked Tracking |
VWO tracking can be blocked for several reasons, such as:
|
These factors prevent VWO from tracking the visitor to respect their privacy choices. This will result in lower visitor counts in VWO than in tools where the same mechanisms may not block.
Be aware that visitors who opt out or are restricted by privacy mechanisms are intentionally excluded from VWO tracking. |
Need more help?
For further assistance or more information, contact VWO Support.