How do you make the best use of B2B ads data?

2

Data Deluge

Marketing data is a firehose.

​

B2B ads data is no different.

Right Percent | OVERVIEW

3

Managing the Data

The best data to focus on is:

• Actionable
• Connected to revenue

Right Percent | OVERVIEW

The Most Important Data From Each Source

4

I: Data Sources

There are two main sources of B2B ad funnel data:

5

The Ad Platform Layer

Right Percent | OVERVIEW

The CRM/Database Layer

​

This includes:

• Ad spend, which is obviously important.
• Top level metrics, like clicks and impressions (and costs per each).
• Conversions, which are events created by the advertiser to signal important visitor activity in the funnel.

6

The Ad Platform Layer

This is the data that the ad platforms can see and report on.

​

In old school marketing, it was very important to track metrics like cost per thousand impressions (CPM) or cost per click (CPC) to help make micro adjustments to campaigns.

​

In modern direct response marketing, you should almost always ignore these metrics.

​

The ad platform use a black box algorithm to show your ads, and often ads with the highest CPMs and CPCs have the best cost per conversion and sale. So these metrics are at best useless, and at worst will give you misleading insights.

​

Right Percent | OVERVIEW

Top level metrics are misleading

Conversion Events

Conversion events are the key for two reasons:

• They help us humans make good decisions about campaign and ad performance.
• They train the AI to find the right kind of clickers who will generate revenue for your business.

​

​

• The main conversion signal you use should be correlated to downfunnel revenue as much as possible - in other words, it should be a leading indicator of revenue.
• You want an event that happens within 1 day of click.
• You have to get 30-50 of of these conversion events per week per ad-set.
• Finding the balance between event value and quantity for your budget is key.

​

Besides your main conversion events that you optimize, you should create events for any important activity on your website or app for all of your ad platforms. It’s a lot of work, but well worth it to make sure campaigns are doing their best at driving revenue for your business.

​

​

Right Percent | OVERVIEW

What do you want from your main conversion events?

The bigger the company you are, the tighter you should be with attribution to make sure you’re only counting incremental sales.

​

Large company should use 7 day click.

​

Small companies can consider 7 day click + 1 day view, since the chance of FB taking credit for coincidental conversions is lower.

​

The more you rely on view through attributions, the less reliable the Facebook algorithm is.

​

10

Right Percent | OVERVIEW

Click or View Based Attribution?

11

​

​

​

The Ad Platform Layer

​

• The important data is spend and conversion events.
• It’s good to establish one or more conversion events as leading indicators of revenue.
• It’s better to rely on click based attribution for conversions.

​

​

​

In Summary

The Key Data to Track

​

For sales driven companies:

• Leads
• -> Opportunities
• -> Closed Won Customers

​

For self-serve driven companies:

• Accounts created
• -> Users engaging with product
• -> Users purchasing product.

12

The CRM/Database Layer

This is the data in your CRM and/or Database.

​

The most important part is tracking the journey of a lead.

Data Associated With Each Lead

13

You need the ability to store marketing tracking data associated with every lead in your CRM/Database as they move downfunnel.

Every lead or sign-up should track the:

• Ad
• Ad-set
• Campaign
• Form download source.

​

Of every new visitor from ads.

​

The most obvious way to measure performance from ad campaigns is to track how far leads generated from those campaigns go down the funnel.

​

This means you can compare campaigns by cost per lead, cost per opportunity, and cost per Closed Won, for example.

​

This a great quality measure with two downsides:

​

1. It can take time to get enough downfunnel conversion data, either due to the events being rare or the events taking a long time from lead to sale.

​

• For self serve businesses, sometimes leads that will never become customers can be very engaged with the product, giving misleading signals.

​

14

Right Percent | OVERVIEW

Measuring Lead and Signup Performance Downfunnel

What can we read from the above sample data?

​

They all have the same cost per Closed Won.

​

Campaign B is likely the best, as it generates the most opportunities. But at this volume of data, it’s not statistically significant.

​

What else can help us make decisions with limited funnel data?

​

15

Right Percent | OVERVIEW

Measuring Lead and Signup Performance Downfunnel

​

​

You can solve the above issues by collecting more info about the lead in the sign-up flow or form.

​

How many employees do they have?

What is their budget?

What is their title?

Do they have intent to buy within 3 months?

Etc.

​

It is vital to ask questions early - it lets you judge campaign quality more quickly and flexibly, working alongside the standard metrics of funnel stage.

​

16

Right Percent | OVERVIEW

​

Solving the Downsides: Collect more lead info up front.

How long does it take a lead to become an opportunity? How long do Opps take to become Closed Won? Does this vary by channel?

This is 100% key to understanding your funnel. If your Closed Wons are happening this week, but your time to vintage is 2 months, then you’re seeing the results of ads you ran two months ago.

That’s why you need leading indicators of lead quality.

When you know your Time to Vintage really well, then you can accurately predict future revenue.

• I.E.: “If I have 5 opportunities within 1 week of ad click date, this will vintage to 30 opportunities from those same clicks within 3 months.”

​

Right Percent | OVERVIEW

What do you need to figure out: Time to Vintage

Ideally you want your internal system to track all prospect touches, so you can manually slice and dice the data.

​

So for instance, if you want see customers from Facebook Campaign X, here’s some sample data on how that might look:

​

• 10 customers with first touch attribution
• 7 with last touch
• 12 with either first or last touch
• 16 with any touch

​

As for your source of truth - in mature clients, we generally recommend first touch for prospecting, with partial credit given for any touch before close date. For clients just starting out, we recommend looking at Any Touch to get learnings faster.

​

18

Right Percent | OVERVIEW

First, last or multi-touch attribution.

There’s a lot of vendors out there who say they can analyze all your clicks to give you a big picture look on how you should attribute sales to different channels.

In B2B, there’s almost never enough downfunnel conversion data to make this work, so these tools are not worth setting up.

To analyze multi-channel attribution, the best first step for most companies is exporting a big spreadsheet of all touches for customers that touched paid ads, and then having an analyst break out the most common patterns.

​

Right Percent | OVERVIEW

Multi-channel Attribution Tools

20

​

​

​

The CRM/Database Layer

​

To best understand B2B ads data, you need to:

1. Track how leads move down your funnel.
2. Track leading indicators of lead value.
3. Track the time it takes for prospects to move from one funnel stage to another on average.

​

​

​

In Summary

Using and Analyzing B2B ads Data

21

II: Using Your Data

Keeping Track of Performance

22

If you’re starting a new campaign (first three months), we don’t recommend dashboards.

​

Instead, we recommend weekly updates on the data, with bullet summaries you can pass upstream.

​

For mature accounts, we recommend the same weekly updates + a weekly updated table of performance.

​

• Early campaign performance indicators will change too often as you test conversion events and landing pages.
• You won’t know time to vintage yet, which makes it hard to interpret dashboards.
• Without downfunnel events populating yet, a dashboard focusing on upfunnel events may cause your team to focus on the wrong thing.
• Campaign structure and naming conventions will change over time as you figure out what’s needed for your business.

​

Why Not Dashboards to Start?

What should be included in weekly updates?

Spend, conversions, and cost per different types of conversion by each campaign in the past week.

Commentary on what the current data means in each campaign, and what the next steps should be.

A conversation on what the quality of current sign-ups and leads looks like.

23

Weekly Updates

Right Percent | OVERVIEW

For a mature business, what should you track on a weekly basis?

​

The best measure is a weekly table that includes: Spend, Leading Indicator Conversion, Cost per Leading Indicator Conversion.

​

Where applicable, it should also have downfunnel conversion, but this should be set-up as a vintage measure that automatically updates over time.

​

24

Right Percent | OVERVIEW

Tracking For Mature Ad Accounts

Note on Vintaging:�When you look at the week of 4/8, it says 1 CW now.

​

As leads vintage, 3 weeks later, the CW number might be 5, and true CPCW might be $200 instead of $1000.

​

​

Even for mature accounts, we don’t recommend weekly dashboard charts (as opposed to tables). They’re too noisy on a weekly basis.

​

Instead, a table with weekly results, with a weekly agenda giving it color, seems to be the most efficient format of sharing learnings.

​

We do recommend charts on the monthly/quarterly basis to illustrate broad narratives.

25

Right Percent | OVERVIEW

Dashboarding

26

​

​

​

Using B2B Ads Data

​

At the start, analyze the top and bottom of the funnel data together on a weekly basis. Over time, develop a weekly tracker that keeps track of your most important metrics, while accounting for time for leads to vintage into customers. On a monthly/quarterly basis, you can graph broad trends to get an instant feel for account progress.

​

​

In Summary