Measurement
Part of Why video marketing agencies need a measurement plan from day one
Choosing attribution for video marketing campaigns
Compare platform data, surveys and modelled tools for video campaign decisions, and learn what to ask agencies about attribution reporting.
What to take away
- For weekly creative and audience calls, platform data is better. It arrives while the campaign is live and ties to the placement.
- Agency attribution methods rarely beat that for in-flight work. Most video agencies reuse platform data or client analytics.
- Surveys such as Wyzowl State of Video Marketing, which found 91% of businesses use video as a marketing tool in 2024, and HubSpot State of Marketing give context, not account-level truth.
- For quarterly budget shifts, a modelled tool such as Northbeam, Triple Whale, Rockerbox or Measured can beat both if you have volume.
- Named video agencies to questionWyzowl, Yum Yum Videos, Demo Duck, Skeleton Productions and Vidico.
Three attribution approaches
Agency attribution methods are the reporting process a video agency applies. Some use last click. Some use Google Analytics 4 or the client CRM. Last click is simple but ignores earlier touchpoints, while GA4 reports cross-channel but depends on consent and tagging, and the client CRM holds offline revenue but lags by weeks.
Platform data is what Meta, Google, LinkedIn and YouTube report. It is fast and tied to spend. It cannot see the invoice that clears three weeks later.
Surveys collect aggregate opinions or spend patterns. They help you benchmark, not reconcile your own orders.
What named video agencies publish
Wyzowl publishes the annual State of Video Marketing survey, which is free to download. Yum Yum Videos produces animated explainers and quotes per project. Demo Duck is a Chicago video production company with public explainer case studies.
Skeleton Productions is a UK video production company that quotes per project. Vidico produces video for tech brands and lists project tiers on its site.
None of these five publicly publishes a full attribution formula. So compare their reporting methods, not a hidden model.
The video marketing agencies measurement and reporting pillar lists fields to agree at kick-off, from conversion definitions to reporting cadence.
Platform data versus modelled tools
Meta Ads Manager reports conversions from its pixel and modelled conversions. Google Ads offers data-driven attribution. LinkedIn Campaign Manager tracks conversions but cannot see offline revenue.
Platform Data vs Modelled Tools
Platform data
- Speed
- Fast, in-flight
- Spend tie
- Tied to spend
- Offline revenue
- Cannot see
- Best account
- Small account
- Examples
- Meta, Google, LinkedIn
Modelled tools
- Speed
- Slower, needs volume
- Spend tie
- Uses spend data
- Offline revenue
- Can include
- Best account
- Large cross-channel
- Examples
- Northbeam, Triple Whale
Modelled tools such as Northbeam, Triple Whale, Rockerbox and Measured spread credit across channels using spend and outcome data. They need volume, identity match and clean spend data. As a typical threshold, modelled tools start to beat platform data above 50 conversions a month per channel, or roughly £25,000 a month in cross-channel spend.
For a small account, platform data plus a holdout test beats a weak model. For a large cross-channel account, a modelled tool can beat platform data.
The verdict
Which Attribution Method Fits?
What decision are you making?
Weekly creative and audience calls -> Platform data
Quarterly budget shifts -> Modelled tool if volume
For quarterly budget shifts, a modelled tool is better if you have the volume. Agency attribution methods are only as good as the definition behind them.
Some campaigns use audiences built from brokered data. The ICO guidance on organisations using marketing services of data brokers sets out lawful basis and transparency duties.
How to compare suppliers
Ask each shortlisted agency for one live report. Look for a named conversion, a stated window and a note on what the method cannot see.
Business demography data such as the ONS changes to business statistics shows how many similar firms formed and closed.
The CIM social media and content marketing training catalogue reflects the analytics skills now in demand.
Read the video marketing agencies measurement mistakes guide before approving a reporting template. Most errors sit in the template.
Glossary
- Last click
- full credit to the final tracked touch before a conversion.
- View-through window
- days after an impression when a conversion still counts.
- Modelled attribution
- statistical credit spread across touchpoints.
- Incrementality
- conversions that would not have happened without the campaign.
- Blended acquisition cost
- total marketing spend divided by total new customers.
Common questions
Do we need modelled attribution on a small budget?
Rarely. A team spending £8,000 a month on media usually lacks volume for a stable model. Platform data plus a holdout test answers more.
Can we run platform and modelled figures side by side?
Yes, in separate columns. Keep one for in-flight calls and one for quarterly budget decisions. Label both.
How often should the attribution method change?
At most once a year, and never mid-campaign. Changing the definition resets the trend line.
Who owns the definition, the client or the agency?
The client owns it. The agency applies it and reports variance. A named owner prevents renegotiation when results dip.
