What a customer decision hub contains, what it costs to operate, how it differs from a CDP or a campaign tool, and when a decision layer fits better.
Frequently asked
Related resources
Solutions
See it in the product
The same loops this note describes run 24/7 against your customer base. Watch the workspace decide, experiment and execute 1:1.
Vocabulary
Definitions an answer engine can quote, each one a page of its own.
Compare
Neutral reads on the categories buyers evaluate against a decision layer.
All comparisonsA customer decision hub is a centralized layer that decides, for a given customer at a given moment, which action is the best one to take, and then hands that decision to whatever channel executes it. It is the opposite of a campaign tool: channels stop deciding for themselves and become delivery surfaces for one shared arbitration.
The term was popularized by Pega, and the category has since been claimed by most large marketing suites. The underlying idea is older and vendor-neutral: one place where eligibility, prioritization and arbitration happen, so a customer does not receive four uncoordinated messages from four teams in the same hour.
Next best action is the output. A decision hub is the machinery that produces it. If you are looking for the definition of the output rather than the platform, start with what next best action means.
A CDP assembles and distributes profiles. It answers who the customer is. A decision hub answers what to do about it. The two are complementary, and a CDP is a common input to a hub, but a CDP alone will not arbitrate between a retention save, a cross-sell and a service message competing for the same moment.
A campaign tool executes a planned send to a planned audience. The hub inverts the direction: the customer arrives, the hub decides. Campaign calendars become a source of candidate actions rather than the unit of work.
This is the part vendor material skips. The licence is rarely the expensive line. The operating cost is:
A heavyweight, in-suite decision hub earns its cost when three things are true: the contact centre and branch network are primary channels (real-time inbound arbitration is genuinely hard), regulatory auditability of every decision is mandatory, and the organization already runs a permanent decisioning team.
For most large B2C bases, the constraint is not arbitration. It is learning speed: how many hypotheses the growth team can test per quarter, and how quickly a proven action reaches the whole base. In that situation, replacing the entire stack is the wrong move. The better shape is a decision layer that sits on top of the systems already in place, reads the same warehouse, and writes actions back into the channels the team already uses.
If you are evaluating suite-native options, our comparisons of Salesforce Einstein and Agentforce and Braze AI decisioning cover the same trade-off in vendor-specific terms, as does the Salesforce next best action walkthrough.
Markin is the decision layer, not the suite: it reads your existing data, ranks candidate actions by expected incremental margin, and executes through the channels you already run. See Customer decisioning.
