Applied machine learning and AI · Istanbul · since 2013
From models
to decisions.
Data Department is an Istanbul-based applied machine learning and AI consultancy. Since 2013 we have helped companies in Türkiye, Europe and the US make better forecasting, pricing, planning and operations decisions, using classical machine learning, optimisation and generative AI built into the systems their teams already use.
Reply within two working daysNo-obligation 30-minute callData stays in your environment
Companies we have worked with
lower inventory holding cost
100,000+ spare parts at an automotive distribution group
forecast error, down from 32%
collections forecasting for a finance company's treasury
lower labour cost
at the same service level, shift optimisation for premium restaurant groups
shorter order waiting times
on average, with very short-term sales forecasting for a quick-service chain
Services
What we do
Classical machine learning, operations research and generative AI, weighed equally. We start from the decision, not the model, and we say so when AI is the wrong tool.
Demand forecasting and inventory optimisation
Probabilistic demand forecasts turned into reorder points, safety stock and production plans your planners can act on.
Read more →PricingDynamic pricing for e-commerce and retail
Price recommendations based on measured demand response, set within the margin and brand guardrails you define.
Read more →Customer analyticsChurn prediction and customer analytics
Churn, propensity and uplift models that put retention and cross-sell budget where it changes customer behaviour.
Read more →Computer visionComputer vision for quality inspection
Vision models and software for defect detection, counting and safety, running on your cameras and tested against your own acceptance criteria.
Read more →Generative AIGenerative AI assistants and agents for business
RAG assistants, natural-language database questions, document AI and scoped agents, each measured against an evaluation set.
Read more →OptimisationOptimisation and simulation for operations
Operations research models that find the best plan under your constraints, and simulations that test how a plan holds up under uncertainty.
Read more →AI strategyAI strategy and fractional Chief AI Officer
Senior AI leadership without a full-time hire, covering roadmap, vendor and architecture reviews, governance, training and team building.
Read more →30 minNot sure which fits?
Tell us the decision you want to improve. In a free 30-minute review we tell you which approach fits, or whether AI fits at all.
Get a free review →Industries
Where we help most
We are most useful where a decision is made many times a week and small improvements compound.
AI and machine learning for the automotive sector
Applied AI for automotive distributors, dealer groups and service networks, from spare parts forecasting to workshop and technician planning.
Read more →Restaurants and food serviceAI and machine learning for restaurants and food service
Applied AI for restaurant chains and food service, from 15-minute sales forecasts to staffing, supply planning and order waiting times.
Read more →Financial services and insuranceAI and machine learning for financial services and insurance
Applied AI for finance companies, insurers, payment companies and fintech lenders, from collections forecasting to renewal and credit-risk models.
Read more →Retail and e-commerceAI and machine learning for retail and e-commerce
Applied AI for retailers and online stores, from store-level demand forecasting and dynamic pricing to retention and recommendations.
Read more →Consumer goods and manufacturingAI and machine learning for consumer goods and manufacturing
Applied AI for consumer goods brands and manufacturers, from demand forecasting and outlet performance to quality inspection and document assistants.
Read more →Distribution and logisticsAI and optimisation for distribution and logistics
Applied AI and operations research for distributors, wholesalers and delivery companies, from inventory forecasting to fleet and route optimisation.
Read more →Telecom and networkingAI and machine learning for telecom and networking
Applied AI and simulation for telecom operators and networking companies, from migration programme risk to service and installed-base analytics.
Read more →How we work
From first call to production
AI opportunity review
A senior consultant looks at the decision you want to improve and tells you honestly whether AI fits.
Discovery
We check your data, rank use cases by value and feasibility, and leave you with a clear go or no-go.
Pilot
One use case, in your environment, measured against a business metric agreed on day one.
Production and handover
Integration with your ERP (enterprise resource planning) or MES (manufacturing execution system), monitoring, and a team that can run it.
Case studies
Results we can show
Spare parts inventory forecasting for an automotive distributor
Read the case study →Premium restaurants 5% to 20%lower labour cost at the same service levelRestaurant staffing and shift optimisation for premium restaurant groups
Read the case study →Financial services 32% → 9.5%forecast error (MAPE), before and afterCollections forecasting for a finance company's treasury
Read the case study →Insights
Practical notes on applied AI
How to choose an AI consulting company: 10 questions to ask
A 10-question checklist for comparing AI and machine learning consultancies, with a scoring table and the warning signs to look for.
Read more →AI strategyWhy AI projects stall after the first model, and how to get them used
Most AI projects stop because nobody uses the output, not because the model fails. Four owners and six checks that keep a model in daily use.
Read more →PricingWhat data should a dynamic pricing model use?
A pricing model is only as defensible as its inputs. Use data that describes the market, not data that describes the person.
Read more →Frequently asked questions
What does Data Department do?
Data Department designs and builds machine learning, optimisation and generative AI solutions that improve operational decisions: demand forecasting and inventory, pricing, customer retention, quality inspection, routing and planning, and AI assistants over company data. We also advise leadership teams as a fractional Chief AI Officer.
Which industries do you work with?
Mostly automotive, restaurants and food service, financial services and insurance, retail and e-commerce, consumer goods (including FMCG, fast-moving consumer goods) and manufacturing, distribution and logistics, and telecom. The methods carry across industries; what matters is a recurring decision with data behind it.
How does a project start?
With a free 30-minute AI opportunity review. If there is a fit, we run a 2-week fixed-scope discovery, then a 6–10-week pilot measured against a business metric, then production and handover.
Do you work with our IT team and existing systems?
Yes. We deploy on your cloud (AWS, Google Cloud, Azure) or on-premises and integrate with your ERP, MES or data warehouse. Your IT team keeps ownership; we document and hand over.
Do you install cameras, sensors or other hardware?
No. Hardware is out of scope. We build the models and software and integrate them with the equipment and systems you already have or choose.
How do you handle data protection (KVKK and GDPR)?
Work happens in your environment under a non-disclosure agreement, and nothing is reused across clients. For generative AI we can use EU-hosted or self-hosted models when data must not leave your control.
Not sure where AI would pay off? Ask us.
Book a free 30-minute AI opportunity review. A senior consultant looks at the decision you want to improve and tells you honestly whether AI fits, what data it needs and what a first step would look like. We reply within two working days.