Definition
Every tracked conversion has a path: the ordered list of interactions, such as visits by channel, ad clicks or email clicks, that the analytics tool recorded for that person before the conversion within its lookback period. The last of those interactions is the one a last-interaction report credits. Every other channel on the path is an assisting channel, and the conversion counts once as an assisted conversion for each of them.
Conversions whose path included the channel before the final interaction(count)
- Conversions are the key events or goals being studied, counted over a stated period. A purchase, a lead and a sign-up should be read separately, because their paths differ in length.
- The path is the sequence of interactions the tool could observe and join to the same person or device within the lookback window. Anything it could not observe is absent, not neutral.
- Before the final interaction excludes the closing touch. A channel that appears several times in one path, including at the end, is usually counted as both the closer and an assister, depending on how the tool defines it, so the definition in use has to be written down.
A companion figure is the assist ratio: assisted conversions for a channel divided by the conversions it closed. A ratio well above one marks a channel that mostly opens or supports journeys; a ratio well below one marks a channel that mostly finishes them. The ratio is a description of role, not a verdict on value.
Older analytics products showed assisted conversions as a fixed column. Current tools tend to show the same information through conversion path reports and attribution comparisons instead, and the names and options change over time, so the current Google Analytics help documentation is the reference for where to find it.
How to interpret it
Assisted conversions answer one question well: which channels appear in journeys that end somewhere else? The guide to measuring content performance uses the count for exactly that reason, because articles are read before people act and are rarely the last thing they touch. The same logic applies to organic search, which is often the first visit; the SEO KPIs and reporting guide recommends showing the assisted figure beside the last-click one with the model named. Social programmes face the same gap, and the social media reporting guide lists assists among the conversion measures a social report should carry.
Read the count by channel role rather than by size alone. A channel planned to build interest, such as the work described under the consideration objective, should show assists far outnumbering the conversions it closes. If such a channel shows neither, it is not taking part in buying journeys at all. A capture channel with many assists and few closes may be bringing people back too early in their decision, or may simply be crowding a path that would have converted anyway.
What the count cannot show is cause. An assist means the channel was present, and presence is easy: a retargeting ad served to someone already on their way to buying collects assists without changing anything. Estimating how much each touch changed the outcome is the job of an attribution model, and the data-driven attribution guide explains how a model learns that split from the paths themselves.
A repair workshop keeps its blog
Tarnside Guitar Repairs spends about £400 a month writing two maintenance articles for its blog. Last-interaction reports credit the blog with 6 of the 140 bookings made over a quarter, and the owner proposes cutting it. The conversion paths report for the same quarter shows a blog visit somewhere before the booking in 52 of those 140 paths, most of them ending with a search for the workshop's name a few days later. The blog's assist ratio is therefore 52 to 6, roughly 8.7. Branded search closed 88 bookings but assisted only 9. The workshop now reports both figures side by side every month, keeps the blog, and tests the effect properly by pausing new articles for one region's landing pages for eight weeks.
What moves it
- The lookback window. A longer window lets older touches onto the path, so assists rise without any change in behaviour. Shortening it removes them.
- Tracking coverage. Consent choices, blocked scripts, cross-device journeys and missing campaign tags all shorten recorded paths. A channel whose visits arrive untagged is invisible as an assister.
- Channel grouping. Assists are counted per channel as the tool groups traffic. Splitting or merging channels, or changing tagging conventions, changes the counts without changing the customers.
- Volume in the channel itself. More visits mean more paths to appear on. Assists can rise because a channel was scaled up, not because it became more useful.
- Which interactions count. A path built from clicks and visits differs from one that also records ad views. Where views are included as touches, display and video gain assists that a click-only path would not show, so the report's definition has to travel with the figure.
What good looks like
There is no reliable benchmark for assisted conversions, and the site does not cite one. The count depends on the length of the buying cycle, the number of channels in use, the lookback window and tracking coverage, all of which differ between businesses and between periods. The useful comparisons are internal: the same channel against its own history under the same window and tagging, the assist ratio against the role the channel was planned to play, and the trend after a deliberate change, such as a new content series. Where the decision is large, such as cutting a channel, the assist count should prompt a test rather than settle the argument.
Related metrics
Assisted conversions are a path-level view of AttributionAssigning conversion credit to touchpoints.: a model shares credit across the same paths that assists simply list. The Lookback WindowThe time after an ad interaction in which a conversion can still be credited to it. sets how far back a path reaches, and so how many assists are possible. The idea of channels with different jobs at different stages of the FunnelModel of the customer journey stages. is what makes an assist worth counting in the first place. A View-Through ConversionA conversion credited to an ad that was shown but not clicked, within a set view window. is a close relative from the ad platforms' side: a conversion after an ad was seen but not clicked, which some reports also treat as a supporting touch. The attribution models framework shows how the same paths produce different channel rankings under each model.
Common mistakes
- Adding assists to last-interaction conversions. One conversion then counts once for the closer and again for every assister, and channel totals exceed real conversions.
- Treating presence as contribution. An assist proves the channel was on the path, not that it changed the outcome.
- Comparing counts across different windows or groupings. A change of settings looks like a change in performance.
- Ignoring untracked touches. Word of mouth, offline adverts and untagged links never appear, so assist counts flatter whatever is easiest to track.
- Cutting a channel on last-interaction figures alone. Channels that open journeys look worthless under a last-click view, and their loss shows up later as weaker results elsewhere.
