300+
retail locations onboarded
99.9%
daily pipeline SLA
5 mo
from roadmap item to revenue-generating add-on
The Client
An established point-of-sale software provider serving independent and small-chain grocers across the US. Its POS terminals and back-office tools processed every scan, sale and inventory movement for hundreds of stores — but that data mostly sat in per-store databases, used for little beyond receipts and end-of-day totals.
The Challenge
Store owners kept asking the same questions: What are my best and worst movers? How do margins compare across departments? What did the promo actually do? The vendor's answer was a clunky export feature, and it was starting to lose deals to competitors advertising built-in analytics.
"Customer-facing analytics" had sat on the roadmap for two years. The engineering team was fully occupied with the core POS product, the company had no data engineers on staff, and US market rates for a senior data team put a serious dent in the business case before a line of code was written. The data itself was messy: hundreds of on-premise store databases on different versions, inconsistent product catalogs, and unreliable store connectivity.
What We Did
Dillo assembled a dedicated augmented data pod — two senior data engineers, an analytics engineer, and a BI developer — embedded in the client's product organization and led day-to-day by the client's VP of Product. The pod was selected through Dillo's AI-automated vetting for grocery/retail data experience specifically.
- Built extraction agents and ingestion pipelines tolerant of flaky store connectivity, with automatic backfill when stores came back online.
- Orchestrated ETL with Apache Airflow, landing data in BigQuery with a per-tenant model that keeps each retailer's data isolated.
- Modeled sales, margin, basket and inventory metrics with dbt — tested, documented transformations with data-quality checks that alert before customers see a bad number.
- Delivered embedded Looker dashboards inside the existing back-office web app: department and item performance, margin trends, promo lift, shrink indicators, and comparisons against anonymized peer benchmarks.
- Set up SLA monitoring and on-call runbooks so the daily refresh — the product's core promise — is treated as a production system, not a reporting afterthought.
- Worked with the client's product and sales teams on packaging, helping shape the analytics tier that went to market.
The Results
- The analytics add-on went from roadmap item to revenue-generating product in 5 months, sold as a subscription tier on top of the core POS.
- More than 300 retail locations onboarded in the first year, with onboarding largely automated per store.
- The daily pipeline has held a 99.9% SLA, with data-quality checks catching upstream issues before dashboards ever showed them.
- Sales now demos analytics in every new deal, and the client reports it as a meaningful factor in both win rate and retention conversations.
- The pod continues as the client's standing data team, now extending the platform toward supplier-facing reporting.
Figures reflect outcomes reported for this engagement; they are project results, not audited benchmarks.
Tech Stack
Services Used
IT Staff Augmentation — a dedicated data pod embedded in the client's product organization, on transparent flat-margin pricing.