SocialHub.AI

Consumer World Model (CWM)

What is a Consumer World Model (CWM)?

A Consumer World Model (CWM) is a per-brand model of your members that turns their own history into a usable profile and a prediction of what they may do next, and shows no score when that history is too thin.

500,000 welcome credits with a new CWM account

What you can use today

The standard analysis available today turns member, order, and behavior facts into one customer profile. Predictions come from that customer's own history. Research directions, including the Behavior Model and the Digital Twin, stay separate from that workspace.

FAQ

What is a Consumer World Model?

A Consumer World Model (CWM) is a per-brand model of your members that turns their own history into a usable profile and a prediction of what they may do next, and shows no score when that history is too thin.

What does CWM stand for, and what is it not?

CWM stands for Consumer World Model, SocialHub.AI's per-brand model of member behavior and prediction. It is not a bare acronym, and it is not a generic research world model. SocialHub.AI stays next to the name.

What is a world model for customer taste?

On this page, a world model for customer taste is a Consumer World Model (CWM) of member behavior: what this customer has done, and a prediction of what they may do next, from that customer's own history. It shows no score when the history is too thin. It is not a fashion taste graph.

What is a world model for customer behavior?

A world model for customer behavior is a Consumer World Model (CWM). It predicts what a customer may do next from that customer's own history, and it shows no score when the history is too thin.

How is a Consumer World Model (CWM) different from a generic world model or a subjective world model?

A Consumer World Model (CWM) is a per-brand model of your members. It predicts from that customer's own history and shows no score when the history is too thin. It is not a general-purpose world model, and it is not a subjective world model of taste.

What is the Behavior Model versus the Digital Twin?

The Behavior Model and the Digital Twin are research and advanced directions. They are described separately from the standard analysis available in the CWM workspace today.

When does a Consumer World Model (CWM) refuse to predict?

When that customer's history is too thin, no score is shown. Unavailable signals are reported explicitly. A new account does not contain customer data or predictions.

Next behaviors, purchase windows and demand context

One member prediction returns next behaviors and personal purchase windows, combining consumer signals with brand demand context when the tenant fusion head is ready.

Interactive example

One consumer. One prediction response.

All states and numbers are synthetic examples, not live predictions or workspace readiness. No customer data is sent.

Choose an example state
Partial result

Behavior predictions remain available. Purchase windows stay unavailable until the brand's fusion head is trained and validated.

What could happen next?

One final distribution over this example's four supported events. Next-event probabilities are uncalibrated.

Next eventProbability
Purchase37%
Portal visit28%
Email click21%
Coupon redemption14%
Why is an event missing?

Email sent is an input context event, not a prediction target. Its probability is null. Other unsupported or uncovered events also stay unavailable.

Will they buy within…?

Personal CWM signals + TimesFM brand demand context.

7 days
Unavailable
30 days
Unavailable
90 days
Unavailable

Chance of at least one purchase. These windows are cumulative; do not add them together or to the event distribution.

Use the same result in your workflow

One member request returns distribution and purchaseWindows together.

Explore API / MCP calls
curl --request GET \
  "https://YOUR_API_HOST/api/v2/cwm/members/YOUR_MEMBER_UUID/next-behaviors" \
  --header "Authorization: Bearer YOUR_API_KEY"

Replace the API host, member UUID and API key with your workspace values.

Required access · predictions:read returns behaviors. Also grant metrics:read (or *) to enable purchase windows when the model is ready.

View response excerpt

Synthetic excerpt only. Brand-model responses also include the full event catalog and coverage metadata; public-prior references use a separate six-class vocabulary.

{
  "predictionKind": "next_event",
  "unit": "probability_0_1",
  "status": "degraded",
  "distribution": {
    "status": "ok",
    "vocabulary": "member_event",
    "normalization": "conditional_on_supported_events",
    "calibrated": false,
    "rows": [
      {
        "key": "purchase.completed",
        "label": "Purchase",
        "probability": 0.37,
        "rank": 1,
        "availability": "predicted"
      },
      {
        "key": "portal.visited",
        "label": "Portal visit",
        "probability": 0.28,
        "rank": 2,
        "availability": "predicted"
      },
      {
        "key": "email.clicked",
        "label": "Email clicked",
        "probability": 0.21,
        "rank": 3,
        "availability": "predicted"
      },
      {
        "key": "coupon.redeemed",
        "label": "Coupon redeemed",
        "probability": 0.14,
        "rank": 4,
        "availability": "predicted"
      },
      {
        "key": "email.sent",
        "label": "Email sent",
        "probability": null,
        "rank": null,
        "availability": "context_only"
      }
    ]
  },
  "purchaseWindows": {
    "unit": "probability_0_1",
    "probabilities": {
      "7": null,
      "30": null,
      "90": null
    },
    "calibrated": false,
    "envelope": {
      "status": "unavailable",
      "backend": "cwm-timesfm-2.5",
      "reasonCodes": [
        "fusion_not_trained"
      ]
    }
  },
  "reasonCodes": [
    "fusion_not_trained"
  ]
}

Synthetic data · TimesFM 2.5

Explore daily demand trends

An example forecast from 128 days of synthetic history. Compare the recent pattern with future daily demand and its quantiles.

Forecast horizon

Synthetic demand units

Recent daily average
20.0
Forecast daily average
20.1
Forecast total
604.3
History · recent 28 daysPoint forecast
Historical demand and future point forecasts with q10–q90 quantilesThe navy line shows 28 days of history. The green line shows the selected forecast horizon. Shading shows the q10–q90 quantile range. Use the day slider to inspect each forecast.
Day 1 · 2026-10-05
Point forecast
22.9
q10
22.8
q90
23.0
View daily forecast data
Date (UTC)Point forecastq10–q90
2026-10-0522.922.8–23.0
2026-10-0621.321.2–21.4
2026-10-0718.718.6–18.8
2026-10-0817.117.0–17.1
2026-10-0917.717.6–17.7
2026-10-1020.020.0–20.1
2026-10-1122.322.3–22.4
2026-10-1222.922.9–22.9
2026-10-1321.321.2–21.3
2026-10-1418.718.6–18.7
2026-10-1517.117.0–17.2
2026-10-1617.717.6–17.7
2026-10-1720.019.9–20.1
2026-10-1822.422.3–22.4
2026-10-1922.922.8–23.0
2026-10-2021.321.2–21.4
2026-10-2118.718.6–18.7
2026-10-2217.117.0–17.1
2026-10-2317.717.6–17.7
2026-10-2420.019.9–20.0
2026-10-2522.322.3–22.4
2026-10-2622.922.8–22.9
2026-10-2721.321.2–21.3
2026-10-2818.718.6–18.7
2026-10-2917.117.0–17.2
2026-10-3017.717.6–17.7
2026-10-3120.020.0–20.1
2026-11-0122.422.3–22.4
2026-11-0222.922.8–22.9
2026-11-0321.321.2–21.3
Download example data (JSON) ↓

Synthetic example, generated with TimesFM 2.5 weights. The forecast starts on 2026-10-05 (UTC); dates are fixed for the demo. Quantile coverage is uncalibrated. Values do not represent a customer's demand or individual purchase probabilities.

One complete next-behavior distribution

Receive one final probability distribution and ranking. The evaluated brand model takes priority, with up to 23 covered targets. When it is unavailable, internal public-prior models form a single six-class reference. Unsupported and context-only events remain unavailable.

7 / 30 / 90-day purchase windows

Included automatically in the same member response: a tenant-private lightweight head combines CWM signals with TimesFM brand trends. Ready heads serve directly; missing training or evaluation leaves only the purchase windows unavailable.

Daily brand and category forecasts

Where the TimesFM 2.5 service is configured and daily history is complete, forecast revenue, orders or units sold for up to 90 days, with point estimates and nine quantiles.

How to read these results

Next-event probabilities sum to 100% over supported candidates and are uncalibrated. Purchase windows separately estimate at least one purchase within 7/30/90 days and are cumulative, never normalized with event classes. Elapsed silence is not conditioned in the event vector. The public-prior reference has not earned workspace validation.

Demand quantiles describe aggregate uncertainty; they are not individual purchase intervals and have not been calibrated. Missing data remain unavailable. No accuracy lift is asserted before your own evaluation.

Forecast API and MCP reference →

What you can use in CWM today.

Available in the CWM workspace

Understand customers and predict behavior

Review observed behavior, consumer state and standard-model signals for purchase, long gaps and cart removal. Results identify missing data and the limits of each signal.

Discover and compare audiences

Explore behavioral lookalikes, clusters and anomaly review. Similarity helps you investigate an audience; it is not a promised conversion rate.

Put intelligence in your workflow

Use the CWM console, REST API or MCP from your application or AI client. Scoped keys control access to your workspace.

Usable analysis requires supported data and enough behavioral history. Unavailable signals are reported explicitly; a new account does not contain customer data or predictions.

Illustrative workflow · not a live prediction

From consumer events to a usable signal.

Your input

  • Member records
  • Order and purchase history
  • Supported browsing and behavior events

Imported with your workspace identity and permissions.

What you review

  • Observed facts and data coverage
  • Available consumer signals and their limits
  • Behavioral similarity and audience candidates

Your team or application chooses how to use the result.

Console / REST API / MCP

One workspace. Your way to use it.

Review analysis in the console or let an authorized application or AI client query CWM. Import supported facts, create a scoped key and follow the public reference for available operations.

Read the API & MCP reference →

Prepaid usage

Start with 500,000 welcome credits.

New CWM accounts receive welcome credits. Eligible operations deduct credits at the published rates. Purchase additional credits in your workspace; your balance and usage remain visible there.

Register with email and password or Google, and agree to the CWM user agreement. Service availability is shown during registration.

Try CWM →

Research & advanced model direction

Explore the science behind the model.

The broader CWM direction includes behavioral modeling and digital twins. Research benchmarks, per-brand model validation and simulation are described separately from the standard operations available in the self-service workspace.

CWM / Harness / Loop Marketing

CWM understands and predicts. Harness orchestrates. Loop Marketing delivers results.

Understand and predict · Core

CWM

Consumer World Model

Understand each customer in one profile and predict churn risk, the next purchase, and 12-month value from that customer's own history. No score is shown when the history is too thin.

Explore CWM →

Orchestrate actions · Runtime

Harness

SocialHub.AI Harness

Orchestrate the next action with governed AI employees and skills, inside permissions, budgets, and review rules.

Explore Harness →

Results you measure · Application

Loop Marketing

Measured outcomes on owned channels

Deliver campaigns, loyalty, and measured results on owned channels: email, SMS, in-app, and store. The customer relationship and the data stay with the brand.

Explore Loop Marketing →

Observed outcomes inform the next cycle. Model updates depend on data quality and validation. The customer relationship and the data stay with the brand.

Start from the customer.

CWM

Start with CWM

Create a workspace, explore the console, and connect your customer data. Purchase credits when you need more capacity.

500,000 welcome credits with a new CWM account

  1. 1Create your CWM account
  2. 2Import supported customer data
  3. 3Run your first analysis
Try CWM →

Harness + Loop Marketing

Orchestrate with Harness and Loop Marketing

See how CWM, Harness, and Loop Marketing fit your customer data, owned channels, and the AI assistants your teams already use.

Book a demo →