Retail Observability Platform

Your data is everywhere.
Your answers aren’t.

Every store has margin leaks, shrinkage drift, and promo misfires. Your team doesn’t have the bandwidth to catch them. Ward watches every store, every KPI, every day. Insight cards explain what changed, why, and what to do next.

Built by operators who scaled 400+ locations and needed answers their BI tools couldn’t give them.
app.getward.ai Live demo
Acme Retail @Merchandising: VP Analyst claude-sonnet default
A

Chat

Ask anything. Ward routes to the right agent and returns cited answers.

Why did Store 37 miss target last week?
You · 9:42 AM
Schema Scout · routed to Merchandising Agent

I pulled Store 37’s last 28 days against the chain baseline. Two root causes — both compounding.

SignalFinding
labor_efficiencyRev/labor-hour −22% vs. cluster — staffing mismatch at 11a–1p peak
inventory.freshFresh fill 83% — backroom replenishment lag at 2–4p
promo.liftBOGO crackers cannibalized Brand Y by 28% — net category +6%

Recommend: re-baseline Store 37 schedule against true peak, raise replen window to 1p, and review the BOGO before next cycle.

8 parallel queries 3 sources cited confidence 0.92
Show me how to fix the staffing mismatch.
You · 9:43 AM
Labor Agent · drafting schedule diff
Querying labor_scheduling…
› Ask anything — Ward routes to the right agent. Cmd+K

Dashboards

Pinned views built from saved data-lake queries.

Revenue vs. forecast +4.2% WoW
Gross margin % −3.2pp
Fill rate — fresh 83%
Shrink — West region +0.8pp

Models

Browse, search, and manage data–lake model definitions for your tenant.

Search models…
NameNamespaceVersion
retail_pos_transactionsretail1.0
retail_inventory_snapshotretail1.2
retail_labor_schedulingretail1.0
retail_promo_calendarretail1.1
retail_supplier_performanceretail1.0
sap_inventory_shrinkagesap1.0
ga4_daily_eventsmarketing1.0
meta_ads_ad_levelmarketing1.0

Sources

Connect external systems to the data lake.

Search sources…
NameTypeLast sync
sap_pos_transactionsimport2m ago
sap_inventory_shrinkageimport2m ago
sap_labor_schedulingimport14m ago
retail_inventory_weeklyimport1h ago
retail_google_ads_dailyimport1h ago
retail_meta_ads_dailyimport1h ago
retail_ga4_website_dailyimport1h ago

Architecture

Two ways to connect. Federate against your live systems, or ingest into Ward’s data lake. Toggle below.

Your systems · read-only
SAP Retail
Snowflake
BigQuery
Shopify
Toast POS
Ward Gateway
TLS 1.3 · AES-256
Querying live · data stays put
Federated answers
SELECT * FROM sap.pos
JOIN snow.inventory
WHERE store_id = 37
→ insight cards
Ward Data Lake
→ baselined per store
TLS 1.3 in transit AES-256 at rest Read-only credentials SOC 2 II in progress VPC peering · PrivateLink

Pipelines

Move data from sources into models on a schedule.

NameSourceModelStatusSchedule
sync_sap_pos_transactionssap_pos_transactionspos_transactionsenabledhourly
sync_sap_labor_schedulingsap_labor_schedulinglabor_schedulingenableddaily
sync_sap_inventory_shrinkagesap_inventory_shrinkageinventory_shrinkageenableddaily
sync_retail_inventory_weeklyretail_inventory_weeklyinventory_weeklyenabledweekly
sync_retail_google_ads_dailyretail_google_ads_dailygoogle_ads_dailyenableddaily
sync_retail_ga4_website_dailyretail_ga4_website_dailyga4_website_dailyenableddaily

Streams

Real-time ingestion pipelines.

0events / min
0streams active
0% delivered
  • pos.txn store_037 — basket $42.18
  • inv.move dc_west → store_104
  • labor.clock store_022 shift_start
  • pos.txn store_211 — basket $19.04

Policies

Browse and manage Cedar access policies for your tenant.

TLS 1.3 AES-256 Read-only SOC 2 II
Policy IDEffectResources
merch-read-defaultpermitModel::*
finance-read-shrinkagepermitModel::"shrinkage"
vendor-blockedforbidModel::"labor_*"
region-west-onlypermitTenant::"acme"

Entities

Principals and resources referenced by Cedar policies.

Search UIDs…
Entity UIDTypeTenant
Tenant::"acme"Tenantacme
Model::"sap.pos_transactions"Modelacme
Model::"sap.inventory_shrinkage"Modelacme
Model::"sap.labor_scheduling"Modelacme
Model::"retail.toast_pos_daily"Modelacme
Model::"retail.ga4_website_daily"Modelacme

Providers

Manage LLM API keys and the model profiles that use them.

API Keys Model Profiles
NameProviderUsed byCreated
anthropic-defaultAnthropic3 profilesApr 22
openai-defaultOpenAI2 profilesApr 22
gemini-defaultGemini1 profileApr 22
ollama-onpremOllama2 profilesApr 22

LLM-agnostic. Bring your own key, route per task. No lock-in.

Settings

Manage your dashboard preferences and account.

Appearance
Theme • Light ° Dark

Light and dark themes are available. Your choice is remembered per browser.

Account
NameAdmin
Emailadmin@acme.io
Tenantacme-retail
Live product. Click any panel — Chat, Sources, Policies — to explore.
SOC 2 Type II in progress TLS 1.3 in transit AES-256 at rest Read-only access
01
Morning Brief

6:47 AM. Problems found
before your team clocks in.

02
How It Works

Stop finding problems
in the post-mortem.

Data In
W
Ward
Insight Out
Action
Feedback

POS, inventory, labor, finance, weather, events. All read into Ward.

Chartered agents query each source. Forecasts fit on the same models your planners trust: ARIMA, Holt-Winters, Bayesian hierarchical, gradient-boosted residuals.

Insight cards arrive. What changed. Why. What to do. Click any number to see the SQL.

Your team acts. Pricing adjusted, inventory repositioned, schedules corrected.

Results feed back. Every outcome tunes Ward. Each cycle is tighter than the last.

A closed loop. Data in, insight cards out, your team acts.

01
Connect the systems your team already uses
POS, ERP, inventory, labor, finance. The same systems your analysts spend weeks reconciling by hand. Ward reads them all. Read-only access. Federated query, data stays put.
TLS 1.3 encrypted
02
Ward baselines every store
Revenue, margin, shrinkage, fill rate, labor efficiency. Baselined per store, per category. Forecasts use the standards your planners already trust. Slow-bleed problems your aggregate reports miss get surfaced one by one.
First insights in 48 hours
03
Insight cards arrive. You act.
Morning brief at 6:47 AM. Anomaly cards through the day. Each card explains what changed, why, and what to do. Click any number to see the model, the SQL, and the source tables behind it.
Lane assist, not autopilot
03
Trust & Governance

Outcomes matter.
So does keeping IT out of trouble.

Most retail AI gets stuck in procurement. Read-write agents pointed at production data. No audit trail. No policy layer. Nobody on the IT side will sign off. Ward is built so your security review is short, your data team sleeps at night, and your CFO sees a cyber liability policy on file before they sign.

SOC 2 Type II TLS 1.3 AES-256 VPC · PrivateLink Read-only by default
Read-only by default
Ward queries your systems with read-only credentials. No writes to your warehouse, your POS, your ERP. Agents cannot mutate the data they reason over.
Federated query · data stays put
Charters on file for every agent
Each agent ships with a written charter: scope, sources, allowed actions, owner, version. Audit it like a job description. Diff it on every change. Sign it off in your ticketing system.
Charter v3 · signed by owner
Agents on a leash
Every agent runs under Cedar policies. Scoped per role, per tenant, per resource. Finance Agent cannot see labor schedules. Vendor Agent cannot touch shrinkage. Least-privilege, machine-enforced.
Cedar policy engine
Every query inspectable
Full reasoning trail on every insight. Which agent ran, which sources it touched, which SQL it issued, which model and which forecast it used. Click any number to see the math. Hand the trail to compliance without a forensics project.
Audit log · reasoning graph
Forecasts your team already trusts
Demand and inventory math runs on ARIMA, Holt-Winters, Bayesian hierarchical, and gradient-boosted residuals. Backtested against your last 24 months before any agent puts a number in front of an operator. The model card travels with the answer.
MAPE on file · backtest report
Bring your own keys
Anthropic, OpenAI, Gemini, Ollama on-prem. Routed per task. Your prompts, your data, your contract. No model trains on your retailer data. Switch providers without re-platforming.
LLM-agnostic · BYO API key
Deploy where IT wants
VPC peering, PrivateLink, or single-tenant in your AWS or Azure account. Your security review covers one network boundary, not a sprawl of SaaS endpoints. SSO and SCIM ship on day one.
VPC · PrivateLink · SSO
Cyber liability covered
Cyber and tech E&O policy on file with an AI rider. Certificate of insurance to your procurement team in a day. Coverage names data breach, regulatory response, and AI-specific liability. Your CFO can stop holding up the redlines.
Cyber · tech E&O · AI rider
Procurement-ready
SOC 2 Type II in progress. Completed security questionnaire on file. DPIA template ready. MSA and DPA in plain English. Ward gets through your vendor review in days, not quarters.
SOC 2 II · DPA · questionnaire
See the security architecture →
04
For Procurement & IT

The directive came down.
Now you’re on the hook.

Leadership wants AI in production by quarter-end. You inherit the integration, the security review, and the blame if it touches the wrong table. Ward gives you the policy plane, the kill switch, and the audit trail your team would build for itself if it had a quarter to do it. AWS-grade governance, without the AWS-grade team. Visualized in a console your IT lead can hand to procurement.

policies/finance-agent.cedar v3 · signed 
// Finance Agent: read-only on warehouse.finance,
// no PII, region-scoped, writes require approval.

permit (
  principal in Role::"FinanceAgent",
  action    in [Action::"read", Action::"summarize"],
  resource  in Source::"warehouse.finance"
)
when {
  resource.classification != "pii"
  && context.region        == principal.region
  && context.budget.tokens > 1000
};

forbid (
  principal,
  action == Action::"write",
  resource
)
unless {
  context.approval.status   == "granted"
  && context.approval.approver in Role::"FinanceLead"
};
Policies live in your repo. Versioned, reviewed, signed.
  • Policy as code, visualized
    Cedar policies live in your Git. Versioned, peer-reviewed, signed. The console renders every rule as a graph: agent, action, source, condition. Your IT team reads policy without reading Cedar. If it’s not in the repo, it’s not in production. No click-admin drift between environments.
  • Scoped per role, tenant, resource
    Finance Agent cannot see labor schedules. Vendor Agent cannot touch shrinkage. US tenant cannot query EU tables. Least-privilege, machine-enforced.
  • Classifications drive access
    Tag a column pii, financial, or operational once. Every agent, every query inherits the rule. No copies of the policy to keep in sync.
  • Who changed what, on the record
    Every charter edit, prompt change, and policy update is logged with name, time, ticket, and approver. Diff two versions side by side. Roll back in one click. Export to your SIEM.
  • Writes need a named human
    Read-only by default. Exports, write-backs, schedule pushes. Anything that mutates state gates on approval from a role you define. Logged with the approver’s identity.
  • Audit any number on the page
    Click a forecast, a margin call, a shrinkage flag. Ward shows the SQL, the source tables, the model, the parameters, and the backtest. Procurement stops asking how the number was made.
Pause
One toggle freezes every agent across every tenant. Resumes clean. Nothing replays.
Kill switch · instant
Audit log
Every prompt, query, model call, charter edit, and answer. Streamed to your SIEM the moment it happens.
JSONL · Splunk · Datadog
Region & tenancy
US stays US. EU stays EU. Single-tenant in your AWS or Azure account on request.
VPC · data residency
Customer-managed keys
KMS or HSM-backed. Rotate, revoke, hold the receipts. Ward never sees the key material.
BYOK · CMK · envelope
Cyber liability policy
Cyber and tech E&O on file with an AI rider. Certificate of insurance to your procurement team in a day.
Cyber · tech E&O · AI
Model card on every number
ARIMA, Holt-Winters, Bayesian hierarchical, gradient-boosted residuals. The forecast, the MAPE, the backtest, on the card.
Forecasts your planners trust
See the architecture → Book a security review →
05
Retail Verticals

Every vertical has problems
hiding in plain sight.

Shrinkage, promo misfires, markdown spirals, demand misreads. The data exists. The answers don’t surface on their own. Ward’s agents bring proven forecasting math to each one. The model that ran is on the card.

30,000+ SKUs
Grocery & Supermarket
U.S. grocery shrinkage topped $37 billion in 2023, with perishable waste accounting for nearly a third.
30K+SKUs per store
15,000+ SKUs
Fashion & Apparel
The average fashion retailer marks down 30–40% of inventory, erasing up to $150 billion in margin industry-wide each year.
15K+SKUs per location
3,000+ SKUs
Convenience & C-Store
Inside-store sales account for only 33% of c-store revenue but 70% of gross profit. Optimizing that mix is the margin game.
24/7daypart monitoring
20,000+ SKUs
Pharmacy & Health
Front-of-store categories drive up to 65% of pharmacy gross margin, yet most chains still manage them with spreadsheet-era tools.
48hrillness demand lead time
50,000+ SKUs
Home Improvement
Home improvement retailers lose an estimated 4–8% of revenue to seasonal overstock and stockout imbalances each year.
50K+SKUs optimized
10,000+ SKUs
Furniture Manufacturing & Retail
Furniture manufacturers operate on 3–6% net margins. A single quarter of misread demand or buried inventory data costs more than most analytics budgets.
3–5disconnected systems unified
5,000+ SKUs
Specialty Retail
Specialty retailers with strong customer data outperform peers by 2–3x on repeat purchase rate and lifetime value.
5K+SKUs per boutique

We ran 400+ locations.
We know what goes unnoticed.

LeanBox. Micromarkets in offices, hospitals, universities. 8-figure revenue, 5 disconnected systems, and problems we found weeks too late. Shrinkage we couldn’t attribute. Promos we couldn’t score. Stores that underperformed quietly. Ward is the platform we built because nothing else caught what we were missing.

Read our story →
06
FAQ

Frequently asked questions.

SAP, Oracle Retail, Shopify, BigQuery, Snowflake, flat files, and any system with a REST API.

First cards within 48 hours. Robust baselines in roughly 2 weeks.

No. Ward sits on top as the intelligence layer that watches your data.

TLS 1.3, AES-256 at rest. SOC 2 Type II in progress. On-prem available.

Yes. Ward scales from 5 stores to 5,000.

Based on store count and data volume. POC engagements at a fixed fee.

Yes. OpenAI, Anthropic, Gemini, or bring your own. Ward routes each task to the optimal model automatically.

See what your stores are hiding.

Margin leaks, shrinkage patterns, promo misfires — the problems are in your data right now. Ward finds them and explains why. First insight cards in 48 hours.

Get a demo →
Get started

Find out what your data has been hiding.

Tell us about your operation. We’ll show you the problems Ward catches — and the ones your current tools miss.

Step 1 of 3
What are your goals?
Step 2 of 3
About your operation
Step 3 of 3
Your contact info