A set of rules that determines how credit for a conversion or goal completion is distributed across the touchpoints a user encountered along their search journey.
SEO practitioners who analyze search performance data using tools like Google Analytics and Google Search Console.
01What it is and how it works
Attribution models assign fractional or full credit to individual touchpoints in the conversion path. In the context of search engine optimization, those touchpoints are organic clicks from search results. The simplest model — last-click — gives 100 percent credit to the final click before the conversion. A first-click model gives all credit to the initial interaction. More nuanced models include linear (equal credit to each touchpoint), time-decay (closer clicks get more credit), position-based (40 percent to first and last, 20 percent to middle), and data-driven (credit is distributed algorithmically using historical conversion data). These rules are applied by analytics platforms such as Google Analytics. For organic search, the model interprets data from Google Search Console and Google Analytics to attribute goals like newsletter signups or purchases to specific queries and landing pages. The mechanism relies on tracking cookies, user identifiers, and a defined conversion window. When a user searches multiple times before converting, the model records each organic click and later distributes credit according to the chosen rule.
It is a way to give fair credit to each organic search interaction that led to a conversion, instead of only rewarding the last click.
02What to do about it
Start by setting up at least one goal in Google Analytics (e.g., a purchase or form submission) and enabling Enhanced Ecommerce if applicable. Then open the Model Comparison Tool under Conversions > Attribution > Model Comparison. Select your goal and compare the default last-click model with a linear or position-based model. Look for keywords that gain value in non-last-click models — these are assisting terms you may underinvest in. Adjust your SEO content strategy: produce top-of-funnel articles for high-assist keywords, even if they rarely end a journey. Conversely, optimize conversion pages for high-last-click keywords. Use UTM parameters on paid campaigns to avoid mixing organic credit with paid. If your analytics supports data-driven attribution, enable it only after you have accumulated at least a few hundred conversions per month; otherwise, stick with rule-based models.
03How it is measured or noticed
You notice the effect of an attribution model by looking at the conversion credit assigned to each organic channel, landing page, or keyword across different models. In Google Analytics, the Model Comparison Tool shows the absolute and relative change in conversions and conversion value for each model. For example, a non-brand keyword might contribute zero conversions under last-click but several under linear. You can also look at the Assisted Conversions report (Google Analytics legacy version) to see which channels appeared in conversion paths without being the last click. In Google Search Console, attribution is not directly modeled; instead, you export click and impression data and join it with conversion data from your CRM or analytics tool to manually compute credits. To notice a discrepancy, run reports filtered by conversion date and look for keywords that appear in assisted positions frequently.
04Common mistakes
- Relying exclusively on the last-click model and ignoring organic interactions that assisted earlier in the journey.
- Applying the same attribution model to all business types regardless of sales cycle length; short cycles need different rules than long research cycles.
- Not segmenting organic traffic from direct, referral, or paid traffic before attributing conversions.
- Assuming that all organic clicks carry equal weight, when time-decay or position-based models are often more realistic.
- Using a default 30-day attribution window when your audience often researches for weeks before converting.
05Limits
Attribution models do not apply well when user identity cannot be tracked across sessions (e.g., after iOS 14 privacy changes or due to cookie consent restrictions). They also break down when a conversion path spans multiple devices and the user is not logged in. Many models ignore touchpoints that occur off-site, such as email or social media engagement, even when those channels influence the first organic search. For branded search, attribution differences between models are often small, so the effort of switching may not be worthwhile. Finally, data-driven models require large volumes of conversion data; with fewer than a few hundred conversions per month, the algorithm may produce unreliable or misleading credit splits.
06A worked example
A user searches 'best hiking boots' and clicks an organic result from your blog. A week later they search 'hiking boots review' and click another organic page on your site. Three days later they search 'buy hiking boots size 10', click a paid ad, and purchase. Under last-click attribution, the paid ad receives 100 percent of the credit. Under first-click, your blog post on 'best hiking boots' gets all the credit. Under linear, each of the three clicks gets 33.3 percent. Under position-based, the first and last clicks get 40 percent each, and the middle click gets 20 percent. The SEO team, looking only at last-click, sees no organic contribution. But with a linear or position-based model, they see that organic content drove 60 to 100 percent of the value. This shifts budget decisions toward maintaining those informational posts.
Frequently asked questions
How is an attribution model different from last-click attribution?
Last-click attribution gives 100% of the credit to the final touchpoint before a conversion, while attribution models distribute credit across multiple touchpoints along the user journey. For SEO, last-click often undervalues organic discovery or early research clicks. Multi-touch models like linear, time decay, or position-based provide a more balanced view of how organic channels contribute.
Should I use a single attribution model or compare multiple?
You should compare several models to understand how credit shifts between channels. Relying on a single model may give a misleading impression of organic performance. For example, if your goal is to show the impact of early research, a first-click or linear model better highlights organic than last-click would.
How do I set up an attribution model in Google Analytics?
In Google Analytics, go to Conversions > Attribution > Model Comparison Tool to view credit distribution across different models. You do not need to change your default model to analyze alternatives. For custom models, use the Attribution Modeling tool in Google Analytics 360 or set up custom channel groupings.
Do attribution models still work after iOS 14 and cookie consent restrictions?
Attribution models become less reliable when user identity is fragmented across sessions or devices, which is common after iOS 14 privacy changes and cookie consent. They still work for first-party data or logged-in users, but for anonymous traffic you may see gaps. Supplement model data with last-click or view-through metrics to maintain confidence.
What are common mistakes when using attribution models?
Common mistakes include relying on a single model without checking others, ignoring assisted conversions, and applying models to channels where user paths are not fully tracked. Another mistake is treating model results as absolute truth instead of directional insight. Always validate with real customer journey data or surveys.
How long does it take to see meaningful data from an attribution model?
It depends on your conversion volume and average time to conversion. For B2B with long sales cycles, you may need three to six months. For e-commerce with daily conversions, two to four weeks can be actionable. Meanwhile, track assisted conversions and path length as leading indicators.
Asked out loud
spoken, not typedThe same term in the words somebody uses speaking to an assistant rather than typing into a box — written from the situation, which is why each one carries the situation it came from.
You might be looking at a last-click model that ignores assists from organic. Try switching to a linear or time-decay attribution model in your analytics tool to see if organic actually contributed to conversions earlier in the journey.
It's a sign you're using last-click attribution. Direct often grabs credit because it's the final click, but organic may have started the journey. Change your default model to first‑click or position‑based in the attribution report to see organic's real value.
Use an attribution model that distributes credit across multiple touchpoints, like time‑decay or linear. That will show you which keywords assisted earlier in the path, not just the final one. Set up a comparison in your analytics tool to present both views.