ChirplyHow can you verify an attribution report?
Start with one conversion ledger, one value per conversion and a documented touch order. Apply each attribution rule to the same rows, then reconcile every model's channel total to the ledger total. A change in channel credit should change the allocation, while the total conversion value stays constant.
An agency analyst needs to explain why organic search receives different credit in three reports. This exercise isolates the allocation rule with four invented journeys. It gives the analyst a reproducible method before investigating identity, lookback windows or provider differences in a real report.
Dataset and download
Download example CSVOpen the CSV in a spreadsheet and keep the original as a reference. Make a working copy for your own review. Downloading saves a local file; it leaves your CRM and communication settings unchanged.
| conversion_id | ordered_channels | conversion_value | currency | data_type |
|---|---|---|---|---|
| EX-1 | organic > email | 120 | illustrative units | synthetic |
| EX-2 | paid search > organic > direct | 240 | illustrative units | synthetic |
| EX-3 | referral | 90 | illustrative units | synthetic |
| EX-4 | organic > organic > email | 150 | illustrative units | synthetic |
Field definitions
- conversion_id
- A unique teaching identifier for one completed conversion; count each once.
- ordered_channels
- Recorded touches from earliest to latest. Direct and repeated channels remain in this exercise.
- conversion_value / currency
- Value assigned to the conversion. Every row uses the same illustrative unit; real reports must use a consistent currency and value definition.
- data_type
- Synthetic means authored demonstration data, not observed customer performance.
Reproduce the method
1. Reconcile the ledger first
Download the CSV and confirm that four unique conversion IDs total 600 units. Keep refunds, cancellations and repeat purchases under a written rule in a real ledger. Do this before grouping by channel so a duplicated conversion cannot disappear inside an aggregate.
2. Calculate first and last touch
For each row, assign the entire value to the first channel, then repeat in a separate column for the last channel. EX-2 assigns 240 to paid search under first touch and 240 to direct under last touch. The exercise deliberately retains direct visits.
3. Calculate linear credit
Divide each conversion value by its touch count. EX-4 has three touches worth 50 each: organic gets 100 and email gets 50. Add each channel across all four paths. Keep full precision until display; rounding each intermediate share can create small differences.
4. Reproduce and explain
Open the attribution calculator. Its Reset example loads these exact paths. Select Calculate results, compare the channel table with the expected results, then Download results CSV. Write the selected model, observation window and identity rule beside the report. Explain the allocation change before recommending a budget change.
Expected results
- All three models allocate exactly 600 units before display rounding.
- First touch: organic 270; paid search 240; referral 90. Last touch: email 270; direct 240; referral 90.
- Linear: organic 240; email 110; paid search 80; direct 80; referral 90.
Interpret the result
Organic receives 270 first-touch units and 240 linear units in this example. That difference describes the rule applied to these paths, not organic's causal contribution.
Google Analytics applies its own channel eligibility, direct-traffic rules and dimension scopes. Use its current documentation when reconciling a GA4 report; this teaching calculator does not reproduce data-driven attribution.
A useful reconciliation compares conversion IDs, timezones, lookback windows, value definitions and missing journeys before comparing model totals. Attribution credit and collected revenue are separate reporting concepts.
Troubleshooting
The model totals differ
Check duplicate conversion IDs, filtered rows, currencies, excluded direct touches and rounding. Reconcile the original ledger total before changing channel labels.
A provider report differs from the example
Confirm its model, direct-traffic behavior and whether the dimension is first-user, session or event scoped. Record the difference explicitly rather than forcing two different definitions to match.
If the CSV does not appear immediately, check your browser's downloads list. Import it as UTF-8 with comma-separated fields if your spreadsheet opens it in one column.
Questions about this example
Does linear attribution mean every channel caused an equal share?
Linear attribution assigns an equal rule-based share to every recorded touch. A causal claim requires different evidence, such as a well-designed experiment.
Can I use these models for phone-call journeys?
Yes, when your ledger reliably connects the observed call touch to a conversion. Preserve unmatched calls as unmatched and state the matching rule; a phone number alone does not establish every touch in a person's journey.
Sources and next steps
- Run the attribution calculator
- Set up call tracking
- Marketing attribution definition
- Google Analytics attribution models
- Google traffic-source dimension scope
The methodology and synthetic rows were authored for this resource and checked against the linked definitions and product guides. Product setup and provider conventions remain documented at their sources. Cite this page's address and review date when sharing the exercise; keep the synthetic-data label with copied rows.