This is a continued post of my thoughts on the topic of Digital Marketing Analytics, specializing in B2B advertising analytics.
What is Advertising Analytics?
To begin with, what is advertising?
Advertising is a marketing tactic to pay for space to promote products and services with the ultimate goal of improving sales.
The purpose of advertising analytics is to determine the return of investment of that tactic – i.e, drawing statistical conclusions on the relationship between the investment and sales.
Advertising Analytics is its own beast because of the large dollar investment in this industry and the huge complexity of the datasets involved.
This will ensure the advertising investment will get a decent return with speed-to-execution and sufficient efficiency analysis, depending on expertise and bandwidth.
Challenges in Advertising Analytics
A foundational challenge in advertising analytics is multi-touch attribution with cross-device adjustments.
No company has solved the challenges perfectly. The complexity of cookie compliance and rapidly changing technology landscape makes this even harder.
Nonetheless, the process to try to address to uncover insights on return remains.
In advertising analytics, the 5 steps of digital analytics are complicated due to:
- # of inputs, including # platforms (i.e, Google, Linkedin, FB..etc) and technologies involved (i.e, Marketo, Rollworks..etc). Not the mention the wide range of offline advertising (e.g, billboard, prints, radio/podcast, TV..etc) mixing with online mediums.
- # of outputs, including # of additional metrics related to #1, plus impression, clicks, CTR and any cost related metrics are additional to the metrics we can measure on web analytics in GA360.
The function of advertising analytics is to draw relationships between those inputs and outputs to see what is working and what is not working.
With enough data and analytical capabilities, the relationship between the two could even be predictable within a range.
I’ll continue to illustrate my thoughts in terms of outputs and inputs, below.
Outputs are the results of digital advertising campaigns, mainly in quantifiable measures and sometimes in qualitative feedback.
In eCommerce, the return of advertising is revenue.
However, in B2B businesses, pipeline is often the quantifiable advertising return, driven from SAO, SAL, MQL and all the way from impressions.
Pipeline is often the north star because it is easier to measure and isolate marketing efforts and effectiveness, rather than direct revenue (which is heavily influenced by sales team effectiveness).
If we go further TOFU, outputs could be extended to include web visits, clicks and impressions.
That highlights why digital analytics is super complex, because the relationship between each stage/step changes from 1:1 (i.e, SAL to SAO) to many:1 (i.e, many web visits to MQL) to many to many (i.e, many impressions to many web visits).
All of this boils down to the fact that, unlike B2C settings, B2B businesses are ultimately selling to companies (i.e, accounts).
And because companies (aka accounts) are comprised of many people, it makes the sales/marketing funnel much more complicated, which in turns make the B2B attribution more difficult.
To complicate the matter further, we have not even scratched the surface of engagement.
Engagement is another way to measure an output that exists somewhere between true TOFU and a strict output – engagement is a signal that a humans are interested in your offering, even if they haven’t actually purchased yet.
If you can analytically measure the signals, that helps your business stay ahead of the competition.
In short, measuring outputs can be summarized as:
- Ad-level awareness metrics: Impressions, click, CTR
- Ad>Web-level engagement metrics: Engaged Users
- “Demand” Funnel: SAO, SAL, MQL
No matter what your business chooses to measure, the key is to have the data available to reflect your effort/activities. If no data is available , you cannot measure at all.
Advertising inputs are a mix of media spend and effort.
- Media spend could be a top-down budget allocation based on the total marketing channel mix.
- Effort is a function of time and headcount required to produce creative and configurations of campaigns.
With every dollar invested, the output should be measured across media spend and effort.
Measuring media spend across following dimensions:
- Ads Platforms (Google, Linkedin, FB…etc)
- Ads Channel (Search, Display, Direct Response…etc)
- Messaging (Go-to-market Themes, Product…etc)
- Location (Geo, Country, Region…etc)
- Segment (Company size, vertical..etc)
- Individual FTE
- Budget Type:
- Always on
- Business Units (e.g, Geo, Product, and Campaign)
Measuring effort is more an art form because it is the management of human resources.
IMHO, the art form is a representation of how well does the advertising manage know about the interest or pain point of potential customers to produce the most effective ads.
Some companies may use a bottom-up approach — the team would need to commit how much output they are going to generate with an ask of the media budget based on the forecast.
Non-technical Challenges in Advertising Analytics
Now that you understand the technical challenges in advertising analytics and how to break down inputs and outputs.
This section talks about the non-technical challenges that relates to people.
Depending on the mature of the business, analytics function can be de-centralized and centralized.
- De-centralized is when the analytics function lives within the advertising team.
- Centralized is when the analytics function lives outside the advertising team, normally in a specalized analytics team.
The decision between de-centralized and centralized is largely driven by how the business processes are defined within the organization.
With talking to many organization on how they structure their teams, it seems to be that the decision of the team structure has no standard or set decision criteria.
It is more of a function of the 1) experience of marketing leadership and 2) existing team know-how.
No matter whether the analytics function is de-centralized or centralized, two things are key to success between ads and analytics teams:
- Clarity of R&R across teams – how do the teams collaborate to address the 5 steps of digital analytics? Are the teams involved resourced (in terms of expertise and bandwidth) to support ads continuously, given ads run 24/7 globally and have a huge spend/impact.
- Expertise in ads analytics – does the team know what needs to be done to be the leader in the space?
In my experience, depending on the maturity of the marketing organization, the in-house advertising team is typically structured with investment breakdown of: 60-70% media spend, 15-20% FTE/agency spend, 15-20% analytics/technology spend.
So here you have it on my brain-dump of digital analytics, and the specialization of web analytics and advertising analytics.
If you are in a startup, it is very likely to be in one big topic of digital analytics.
If you are in a larger B2B organization, specialization of web analytics and advertising analytics can give you some structure to ask the right questions to get the right answers.
At the end of the day, the devil is in the details of execution.
It not only requires technical knowledge of technology but also how to get alignment across teams and the right investment of resources to achieve the desired analytical engine.