Friday, August 19, 2016

Guide to ABM Vendors: What's in a Complete ABM Stack?

Yesterday’s post announced our new Guide to ABM Vendors, which helps marketers make sense of the confusing variety of ABM-related systems. The post describes our framework of four ABM process steps, six system functions that support those steps, and six sub-functions that are hardest to find. These are summarized in the table below:


ABM Process
System Function
Sub-Function
Identify Target Accounts
Assemble Data
External Data
Select Targets
Target Scoring
Plan Interactions
Assemble Messages
Customized Messages
Select Messages
State-Based Flows
Execute Interactions
Deliver Messages
Execution
Analyze Results
Reporting
Result Analysis


You might assume that a complete ABM stack would include systems that do all of the functions and sub-functions, but that’s only half right. All the functions are indeed required, but the sub-functions are optional. Customized Messages and Execution can be found in non-ABM systems such as Web site personalization or marketing automation.* The rest, including External Data, Target Scoring, State-Based Flows, and Result Analysis, make ABM better or easier but you can do ABM without them.

I know you really want to learn which vendors do which functions, but, as I also explained yesterday, that's not a simple question to answer.  Instead, let’s start with an overview of how the vendors as a group matched up against the sub-functions.

The first obvious question is how many vendors deliver each sub-function. The table below gives the answers, and they’re pretty much what you’d expect: lots of vendors use external data and do execution, a fair number who do target scoring and result analysis, and relatively few do customized messages and state-based flows. Bear in mind that customized messages and state-based flows (which are roughly equivalent to multi-step campaigns) are widely available in general purpose systems not covered in the ABM Guide, so the low counts here don't mean they are really hard to find.



Sub-Function
Number of Vendors
External Data
28
Target Scoring
15
Customized Messages
6
State-Based Flows
10
Execution
19
Result Analysis
16
 
The preceding table implies that most vendors deliver multiple sub-functions (otherwise, the numbers would total to 40). In fact, just eleven vendors qualify for a single category. On the other hand, none deliver all six sub-functions and only seven deliver four or five. The table below shows the details. What it means in plain English is that no vendor provides a complete ABM solution and that most are quite specialized. In still plainer language, it means you'll need more than one ABM vendor.



Number of Sub-Functions
Number of Vendors
1
11
2
14
3
8
4
4
5
3
6
0

This is even truer when you recognize that the sub-functions themselves are very broad categories, so vendors who qualify for the same sub-function may be delivering significantly different products. You may not need to catch ‘em all, but you certainly need to work with several.**


Right now, the more analytically inclined among you are probably wondering if there’s any pattern to which sub-functions are provided by the same vendor. There sure is! The table below shows values for individual vendors (names removed), sorted by sub-function.




Identify Target Accounts
Plan Interactions
Execute Interactions
Analyze Results

Assemble Data
Select Targets
Assemble Messages
Select Messages
Deliver Messages
Reporting

External  Data
Target Scoring
Customized Messages
State-Based Flows
Execution
Result Analysis
data only
1





1





1





1





1





1





1





1





data and delivery
1

1

1
1
1



1
1
1



1
1
1



1

1



1

data and target scoring
1
1




1
1




1
1




1
1




1
1




1
1




1
1




1
1




1
1




1
1



data, scoring, and delivery
1
1
1

1
1
1
1

1
1
1
1
1


1
1
1
1


1

custom messages

1
1
1
1
1


1
1
1
1


1
1
1
1


1

1

journey management



1
1
1



1
1
1



1
1
1



1
1




1
1

1


1

1




1

measurement





1





1


As the color coding shows, I see the vendors falling into seven clusters having similar combinations of sub-functions. These are
  • data-only vendors, who accumulate account data to resell to others. They often specialize in particular data types such as intent or technology use. The ABM Guide refers to these as “commercial” data vendors.
  • data and delivery vendors, who accumulate account data and use it to deliver targeted messages. They often specialize in channels such as display ads.
  • data and target scoring vendors, who accumulate account data and use it to build models that select target accounts. Many of these vendors combine data they gather themselves (which they often consider a competitive advantage) with data they purchase from commercial vendors.
  • data, scoring, and delivery vendors, who accumulate data, build targeting models, and deliver messages based on those models. 
  • custom message vendors, who create and deliver messages that are dynamically tailored to specific accounts or segments. Most of these vendors also qualify for the State-Based Flow sub-function because they automatically move accounts into new segments over time. But their focus isn’t on comprehensive journey management and they don’t orchestrate messages delivered by other systems. So they're not quite as powerful as all those checkmarks might suggest.
  • journey management vendors, who don’t create custom messages but do move accounts into new segments based on behaviors and external data. This cluster includes several single-channel vendors, who are really quite different from multi-channel vendors but get checks in the same boxes. Even the true multi-channel journey managers send messages in just a few channels and rely on external delivery systems for the rest.
  • measurement vendors, who specialize assembling account-level data and reporting on it.
You’ll notice the clustering isn’t perfect – there are a few cells with a 1 outside the highlighted areas and a few blank cells within the highlights. There’s nothing wrong with that. In fact, the exceptions are quite interesting although I won’t be discussing them here. My point at the moment is that we’re starting to make sense of the vendor chaos.

The following table collapses each of the seven segments to a single row, with the numbers showing how many vendors in each segment qualified for each sub-function. This shows even more clearly how the pieces of an ABM stack should fit together.




Identify Target Accounts
Plan Interactions
Execute Interactions
Analyze Results

Assemble Data
Select Targets
Assemble Messages
Select Messages
Deliver Messages
Reporting

External  Data
Target Scoring
Customized Messages
State-Based Flows
Execution
Result Analysis
data only
8





data and delivery
5

1

5
3
data and target scoring
10
10




data, scoring, and delivery
4
4
1
4
3
custom messages

1
4
3
4
3
journey management


6
6
4
measurement





2

To put this table in a more narrative format:
  • ABM starts with data, since the goal is to identify target accounts in advance instead of marketing to anyone you happen to reach. Your internal data won’t be adequate so external data is essential. So data from one or more vendor will be the foundation of your ABM effort.
  • You can jump directly from data to messaging if you want to do all the intermediate steps by yourself (target selection, message creation, and journey management). Clearly this is more work but it does lead to simplest ABM stack possible. The data-and-delivery vendors tend to be channel-specific, so you might need more than one and you may struggle to coordinate their campaigns.
  • If you do want help, you can start by hiring a scoring vendor to identify your best target accounts. They’ll usually bring their own data and combine it with your own information.
  • Most scoring vendors simply hand over a scored target list plus maybe some data.  But others add delivery services, providing the most comprehensive single vendor solution available. As with the data-and-delivery vendors, most data-scoring-and-delivery vendors serve a subset of channels, so you’ll need to supplement them with other execution systems.
  • Somebody has to create your messages. If you want advanced customization, you’ll probably need a content building specialized tool. That product won’t necessarily be limited to ABM applications, though, so you’ll have options beyond the vendors listed in the ABM Guide. Most systems that create customized messages can also deliver them, although again you'll need several systems to cover all your channels.
  • Messages need to change as the account moves through its journey. Some of the messaging and delivery products do this within specific channels, which still leaves you to manually coordinate messages across channels. Multi-channel journey managers built especially for ABM are just starting to appear; you can buy one today if you’re willing to be an early adopter. More will show up as vendors from other clusters extend their products and as generic marketing automation systems and journey orchestration engines add features to meet ABM requirements.
  • Most systems that deliver messages also report on the results. Again, the trick is finding a solution that combines data from all channels. If your delivery systems are fragmented, you'll probably need a separate cross-channel reporting system to build a full picture of your ABM program.
So…maybe this clears things up and maybe it leaves you more confused than ever. When you cut through all the pesky details, it comes down to two main points:
  • at a minimum, you need data to pick target accounts and delivery systems to send them messages. Everything in between is optional but helps to make your work easier and more effective.
  • there are a lot of single-channel ABM solutions. There’s nothing wrong with that but remember that you’ll ultimately want to coordinate across channels.  So don’t think one single channel solution can fill all your needs by itself.

Finally, as a reward for reading this far and because you’ve now absorbed enough nuance to understand how tentatively such lists must be read***, here is a table showing which vendors I’ve placed in each cluster:


Cluster
Vendors
data only
Bombora, Data.com, DiscoverOrg, HG Data, InsideView, Orb Intelligence, ReachForce, ZoomInfo
data and delivery
Azalead, LinkedIn, Madison Logic, True Influence, Vendemore
data and target scoring
Avention, Datanyze, Dun & Bradstreet, GrowthIntel, Infer. Lattice Engines, Leadspace, Mintigo, Radius, Everstring
data, scoring, and delivery
Demandbase, Mariana, MRP, The Big Willow
custom messages
Evergage, Kwanzoo, SnapApp, Triblio
journey management
Engagio, GetSmartContent, LookBookHQ, Terminus, Uberflip, YesPath, ZenIQ
measurement
Bizible, LeanData



________________________________________________________________________________

*You could argue that those should be considered ABM systems, especially when they’re used for ABM programs. I might even agree. But I had to draw the line somewhere when deciding which systems to include in the ABM Guide, and there are too many general purpose products to include. That said, marketers looking for ABM solutions should certainly assess whether non-specialist products can meet some portion of their needs. The assessment framework in the ABM Guide applies quite nicely.

**Which in turn implies a PokĂ©mon Go-style vendor collecting game. The really good conference organizers are probably working on this already. I’m looking at you, Scott and Nikki.

***and therefore why you need to buy the Guide to ABM Vendors to make a wise choice.





No comments: