Striped Giraffe Innovation & Strategy GmbH, Lenbachplatz 3, Munich (2026)

10/06/2026

In just 13 months, Data Product adoption jumped from 48% to 69%.
Such rapid growth points to a fundamental shift in how organizations manage and operationalize data.
And according to the latest BARC research, that shift has major implications for AI success.

Organizations that scaled Data Products across the company are 3.4x more likely to move AI into production.

Not pilots. Not demos for board meetings.
Actual production AI.

The global research, conducted by BARC among 300+ data and analytics leaders across 20+ countries, shows how quickly the market is changing:

🔸 85% of companies with enterprise-wide Data Products already run 3+ AI projects in production
🔸 77% already have agentic or autonomous AI systems deployed at least in limited production

What’s particularly interesting is why companies invest in Data Products today.

A few years ago, the dominant narrative was democratization and self-service.
Now the priority is much more pragmatic:

➡️ trustworthy data for AI
➡️ reliable decision-making
➡️ clear ownership and accountability
➡️ governance that survives scale

And honestly? That shift makes sense.

Many organizations discovered the hard way that scaling AI on unstable, poorly governed data creates expensive chaos very quickly. Especially when AI starts acting autonomously inside business processes.

One insight from the report stands out for us in particular:

Regulated industries like insurance and financial services are ahead not because they are “more innovative,” but because years of compliance pressure forced them to build disciplined data operating models earlier than everyone else.

That lesson matters far beyond regulated sectors.

As Łukasz Cempulik, DWH and BI Architect at Striped Giraffe, puts it:

“Companies often approach Data Products as a data architecture initiative. The organizations achieving the strongest outcomes treat them as an operational trust model for AI.
At scale, AI reliability becomes inseparable from data ownership, observability, and accountability embedded directly into business processes.”

After five previous posts in our Data Poducts series, one conclusion becomes difficult to ignore:

The companies winning with AI are increasingly the companies that learned how to operationalize trustworthy data first.

How mature is your organization’s Data Product approach today?
Let us know in the comments.

In the comments, you can find links to all previous posts in the Data Products series.

09/06/2026

Most industrial manufacturers still measure success by how many machines they sell. But the real business increasingly starts 𝙖𝙛𝙩𝙚𝙧 𝙩𝙝𝙚 𝙨𝙖𝙡𝙚. 💡🧐

According to Boston Consulting Group (BCG)’s analysis, aftermarket services often deliver gross margins that are twice the 15% to 25% typically earned from equipment sales. And in many sectors, installed-base monetization is becoming more stable than cyclical CAPEX-driven revenue.

You can see this shift clearly across the industry:

🔸 Caterpillar Inc. generated around $24B in services revenue in 2023, up from approximately $14B in 2016, reflecting the growing importance of lifecycle services and connected equipment strategies.

🔸 Boeing Global Services generated nearly $19B in revenue in 2023, driven by maintenance, parts, digital solutions, and lifecycle support services for the global installed base.

🔸 GE Aerospace reported in 2024 that more than 70% of Commercial Engines & Services revenue was tied to services-related activities.

What’s changing is not just the economics — but the operating model behind it.

Industrial leaders are moving from:
➜ selling equipment and spare parts
to:
➜ managing lifetime customer value

That shift changes everything:
🔹 how installed-base data is managed
🔹 how spare parts are forecasted
🔹 how service operations are prioritized
🔹 how customer relationships are maintained post-sale

And this is where many organizations hit a wall.

Because most OEMs still have fragmented visibility into their installed base:
ERP in one place, service history in another, IoT data somewhere else, local spreadsheets everywhere.

Without connected visibility, proactive aftermarket is almost impossible.

That’s why “industrial aftermarket” is no longer just a service topic.
It’s becoming a data, architecture, and customer experience challenge.

We explored this transformation in our article on industrial aftermarket opportunities:

👉 https://www.striped-giraffe.com/en/blog/industrial-aftermarket-from-spare-parts-business-to-a-scalable-revenue-engine/

09/06/2026

𝗧𝗵𝗲 𝗿𝗲𝗮𝗹 𝘁𝗲𝘀𝘁 𝗼𝗳 𝗮 𝗗𝗮𝘁𝗮 𝗣𝗿𝗼𝗱𝘂𝗰𝘁 𝗶𝘀 𝗵𝗼𝘄 𝗲𝗮𝘀𝗶𝗹𝘆 𝘀𝗼𝗺𝗲𝗼𝗻𝗲 𝗰𝗮𝗻 𝘀𝘁𝗮𝗿𝘁 𝘂𝘀𝗶𝗻𝗴 𝗶𝘁.
In mature organizations, access is designed as a clear, repeatable flow — typically structured around two layers: discovery and delivery.

1. 𝗗𝗜𝗦𝗖𝗢𝗩𝗘𝗥𝗬 𝗟𝗔𝗬𝗘𝗥

Users go to a data marketplace or catalog, which becomes the primary entry point.

This is where Data Products are presented in a structured, business-friendly way:
🔸 product owner
🔸 business description
🔸 data quality indicators
🔸 date of last update
🔸 available access options (e.g. SQL, BI, API)
🔸 clear action: “Request access” or “Get access”

The goal is simple: make data easy to find and understand — without technical knowledge.

2. 𝗗𝗘𝗟𝗜𝗩𝗘𝗥𝗬 𝗟𝗔𝗬𝗘𝗥

Once access is granted, data is typically not “sent” anywhere. Instead, it becomes available in the tools users already work with:

🔸 BI tools / connectors → for business users (reporting, dashboards)
🔸 SQL endpoints → for analysts working on curated datasets
🔸 APIs (REST / GraphQL) → for applications and digital products
🔸 Data sharing (e.g. Snowflake, Delta Sharing) → for cross-team or external access
🔸 Event streams → for real-time operational use cases

In some cases, data can also be delivered as exported datasets (e.g. CSV) available in the data marketplace.

The key principle: access should align with how the data is actually used.

𝗔𝗖𝗖𝗘𝗦𝗦 𝗠𝗢𝗗𝗘𝗟𝗦

Most platforms combine two approaches:
🔸 Instant (policy-based) → access granted automatically based on roles and rules
🔸 Request-based (workflow-driven) → access approved by data owners when needed

𝗪𝗵𝗮𝘁 𝗵𝗮𝗽𝗽𝗲𝗻𝘀 𝗯𝗲𝗵𝗶𝗻𝗱 𝘁𝗵𝗲 𝘀𝗰𝗲𝗻𝗲𝘀

In mature environments, access is automated end-to-end:
1️⃣ User selects a Data Product and access type
2️⃣ Policy engine evaluates permissions (role, sensitivity, purpose)
3️⃣ Access is granted automatically or routed for approval
4️⃣ Roles and permissions are provisioned (e.g. in Snowflake / Databricks)
5️⃣ User receives notification — data is ready in their tool

No tickets. No manual handoffs.

What makes it work

🔸 self-service as default
🔸 clear access paths aligned with users
🔸 policy-based automation (RBAC / ABAC)
🔸 marketplace-style experience
🔸 minutes from discovery to usage

Łukasz Cempulik, DWH and BI Architect at Striped Giraffe:
“Access is where Data Products either scale or quietly fail. If reaching the data requires much effort, users will always find an easier path.”

How long does it take in your organization to go from finding data to actually using it? Let’s discuss!

08/06/2026

𝗧𝗵𝗲 𝗿𝗼𝗹𝗲 𝗼𝗳 𝘁𝗵𝗲 𝗕𝟮𝗕 𝗽𝗼𝗿𝘁𝗮𝗹 𝗶𝘀 𝗾𝘂𝗶𝗲𝘁𝗹𝘆 𝗰𝗵𝗮𝗻𝗴𝗶𝗻𝗴.
For years, portals were built as transaction interfaces:
place an order, download an invoice, track a shipment.

Today, they are becoming operational coordination layers.

Because in modern B2B, processes no longer run in isolation:

➜ A pricing update impacts contracts.
➜ Inventory changes affect delivery commitments.
➜ Service cases influence renewals.
➜ Subscription models require constant alignment between finance, sales, and operations.
➜ Suddenly, the portal is no longer just part of the customer journey.
➜ It becomes part of how the organization actually works.

That changes the architecture behind it.

Modern self-service ecosystems need API-first architecture, event-driven integrations, real-time visibility, role-based workflows, and governed access to data and actions.

The goal is not only better customer interaction.

It is operational continuity across teams, systems, and decisions.

The companies moving fastest reduce the distance between information, decision-making, and ex*****on.

And that directly improves scalability, responsiveness, and organizational speed.

Our latest e-book explores how enterprise self-service ecosystems are evolving into integrated operational platforms.

👉 https://www.striped-giraffe.com/en/e-book-self-service-and-automation-in-b2b-ecommerce/

28/05/2026

Many B2B companies don’t run slowly because their systems are outdated. 🐌
They run slowly because critical dependencies are still hidden inside everyday processes. 🖇️

A sales rep waits for pricing approval.
Customer service waits for logistics updates.
Procurement waits for contract verification.
The customer waits for everyone.

And suddenly, a company with modern systems still operates at the speed of email threads.

This is where many self-service initiatives fall short.

They are treated as commerce features: order tracking, invoice downloads, reorders, FAQs.

Useful? Absolutely.
Transformational? Not even close.

The real shift begins when self-service becomes part of the operating model.

When a key account manager can approve pricing within defined thresholds, check inventory and delivery risks in real time, access customer-specific contracts instantly, and trigger workflows without opening another ticket.

When customers receive proactive alerts about delays, alternative delivery options, or maintenance recommendations before they even contact support.

That changes more than customer experience.

𝗜𝘁 𝗰𝗵𝗮𝗻𝗴𝗲𝘀 𝗼𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝘀𝗽𝗲𝗲𝗱.

Robert Gruca, E-Commerce Solution Consultant at Striped Giraffe:

“We increasingly see that the strongest B2B organizations are reducing operational friction by removing unnecessary coordination layers between people, systems, and decisions. And automation plays a major role here.”

But the strategic objective is broader:

“Creating an organization where employees and customers can act autonomously, with reliable real-time information and clearly governed processes.”

That’s a very different conversation from “digital ordering.”

In our latest e-book, we explore how B2B companies are building self-service ecosystems that combine:
🔸 operational autonomy
🔸 process automation
🔸 personalization
🔸 real-time analytics
🔸 scalable architecture

👉 https://www.striped-giraffe.com/en/e-book-self-service-and-automation-in-b2b-ecommerce/

27/05/2026

Many organizations invest heavily in building Data Products.
Very few measure whether those products actually succeed. 📐🎯

Most teams track pipelines, infrastructure, and system performance.
But a Data Product should be evaluated like any other product: by its adoption and impact.

Three categories of metrics matter most.

1️⃣ 𝗔𝗱𝗼𝗽𝘁𝗶𝗼𝗻 𝗺𝗲𝘁𝗿𝗶𝗰𝘀

First, measure whether the product is actually used.

Examples include:
🔸 number of active users
🔸 number of consuming teams
🔸 API calls or queries
🔸 integrations with other systems

If usage is low, the product likely solves the wrong problem.

2️⃣ 𝗥𝗲𝗹𝗶𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗺𝗲𝘁𝗿𝗶𝗰𝘀

Consumers depend on predictable data.

Key indicators include:
🔸 data freshness
🔸 SLA adherence
🔸 incident frequency
🔸 data quality scores

Without reliability, adoption quickly collapses.

3️⃣ 𝗕𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗶𝗺𝗽𝗮𝗰𝘁 𝗺𝗲𝘁𝗿𝗶𝗰𝘀

Ultimately, Data Products must improve decision-making.

Organizations often track outcomes such as:
🔸 faster decision cycles
🔸 improved forecast accuracy
🔸 better campaign performance
🔸 reduced operational inefficiencies

This is where data initiatives prove their real value.

Łukasz Cempulik, DWH and BI Architect at Striped Giraffe:

"Treating data as a product means accepting one simple principle: success has to be measurable. If you can’t link a Data Product to real usage and business outcomes, it’s very hard to justify its value.”

How does your organization evaluate the impact of its Data Products?
Let us know in the comments. 💬

21/05/2026

Almost every company says automation issues lead to SLA breaches. 61% experience them at least monthly.
And yet, nearly three-quarters still lack end-to-end visibility into their automated business processes. 🫣😬

That’s not a technology gap; it’s a control problem.

A recent global Broadcom study of 500+ executives and IT leaders reveals a pattern that’s hard to ignore:

More automation ➜ more tools ➜ less clarity.

What’s really happening inside many organizations:

🔸 80% run 3+ automation platforms
🔸 74% use 3+ monitoring tools
🔸 74% deal with “alert storms” that obscure root causes

Instead of clarity, teams get noise.
Instead of faster resolution, they get longer investigations.

And the consequences are business-critical:

1️⃣ Issues are detected too late
Most teams only react after something fails.

2️⃣ Impact is unclear
68% can’t tell if a problem will breach an SLA.

3️⃣ Customer experience suffers
88% say SLA breaches directly affect customers.

4️⃣ Costs quietly increase
69% lack the data to properly optimize workloads.

The uncomfortable truth:
Automation has scaled faster than your ability to understand it.

You’ve built highly distributed, event-driven systems —
but visibility has not kept up.

So when something breaks, you don’t just fix issues.
First, you search for them.

This is exactly where the real risk sits today.

If you’re interested in how to close this gap — not by adding more tools, but by making processes truly visible and understandable — we explore this in detail in our latest article:

The Visibility Gap in Automation: Where Processes Lose Transparency
🔗‍️ https://www.striped-giraffe.com/en/blog/the-visibility-gap-in-automation-where-processes-lose-transparency/

20/05/2026

“What’s the best medicine for sinus pain during pregnancy?” 🤔❓
Today, that question is more likely to land in Google or ChatGPT than on an online pharmacy platform.
And that should concern the entire e-pharmacy industry.

Most online pharmacies still operate like traditional retail catalogs:
search by product name, brand, or active ingredient.

But healthcare decisions rarely start that way.

Patients usually begin with:
🔸 symptoms
🔸 uncertainty
🔸 restrictions
🔸 side effects
🔸 interactions
🔸 questions about substitutes or safer alternatives

This is why many pharmacy platforms are quietly evolving beyond transactional e-commerce.

Across Europe, leading players increasingly invest in:
🔹 symptom-based discovery
🔹 AI-assisted search
🔹 substitute and equivalent mapping
🔹 structured product data
🔹 richer medical content
🔹 integrations with telehealth and consultation services

Because the real challenge is no longer putting pharmaceutical products online.
The challenge is helping people navigate healthcare decisions safely, clearly, and with confidence.

And this becomes surprisingly complex in e-pharmacy.

A single product may involve:
🧪 dosage variations
📛 contraindications
💸 reimbursement rules
🌍 country-specific restrictions
🔀 interactions with other medications
⚖️ multiple substitutes and equivalents

Without properly structured and connected data, even the best UX quickly breaks down.

That is why search, discovery, and product information architecture are becoming strategic capabilities in digital pharmacy.

Not just for conversion. But for trust.

Want to explore how data management supports digital pharma and healthcare platforms?

Discover key insights in our free e-book “Data Management in Pharma & Healthcare”:
🔗 https://www.striped-giraffe.com/en/e-book-data-management-in-pharma-healthcare/

18/05/2026

Launching financial products is no longer the problem.
Controlling how they perform is.
And that’s exactly where most banks lose scale. 🤔⬇️

This is the problem we address in our latest article — Part 2 of the series:
“From Financial Institution to Commerce Engine.”

The core insight:
The limiting factor is not speed, channels, or customer access.
It is the loss of control at the moment where product logic, customer data, and decisioning are supposed to come together.

𝗪𝗵𝗮𝘁 𝘆𝗼𝘂’𝗹𝗹 𝗳𝗶𝗻𝗱 𝗶𝗻 𝘁𝗵𝗲 𝗮𝗿𝘁𝗶𝗰𝗹𝗲:

1️⃣ The hidden bottleneck: product governance

🔸 Product logic fragmented across systems
🔸 Pricing, eligibility, and bundling misaligned
🔸 No single control point over the offer

The result: multiple versions of the same product in market.

2️⃣ Why customer data still falls short

🔸 Built for segmentation, not decisions
🔸 Distributed across systems with different timing
🔸 Lacking real-time context

Two different customers — same offer.

3️⃣ Where banks lose control

🔸 Decisions precomputed in batches
🔸 Logic embedded across channels
🔸 Exceptions handled manually

This leads to latency, inconsistency, and missed revenue.

4️⃣ What leading banks do differently

They connect three layers into one system:
🔸 Governed product model (what’s possible)
🔸 Decision-ready customer layer (what’s relevant)
🔸 Real-time decisioning (what’s executed)

Not as separate capabilities — but as a single operating model.

5️⃣ The shift that enables scale

From:
🔸 fragmented logic → governed orchestration
🔸 campaign decisions → real-time decisions
🔸 system constraints → controlled flexibility

This is where financial commerce actually starts to scale.

Want to learn more?

👉 𝗥𝗲𝗮𝗱 𝘁𝗵𝗲 𝗳𝘂𝗹𝗹 𝗮𝗿𝘁𝗶𝗰𝗹𝗲 here:
https://www.striped-giraffe.com/en/blog/from-financial-institution-to-commerce-engine-part-2-the-missing-layer-product-and-customer-orchestration/

13/05/2026

The most successful Data Products don’t start with data.
They start with users 🙋🏻 and the decisions 🔀 those users need to make.
𝗧𝗵𝗮𝘁 𝘀𝗵𝗶𝗳𝘁 𝗰𝗵𝗮𝗻𝗴𝗲𝘀 𝗲𝘃𝗲𝗿𝘆𝘁𝗵𝗶𝗻𝗴 — from how the data is structured to how it is delivered, documented, and maintained.

In real projects, effective Data Products follow a very different design logic.

1️⃣ 𝗦𝘁𝗮𝗿𝘁 𝘄𝗶𝘁𝗵 𝗯𝘂𝘀𝗶𝗻𝗲𝘀𝘀 𝗱𝗲𝗰𝗶𝘀𝗶𝗼𝗻𝘀

Before designing a Data Product, ask a simple question:

What decisions should this data support?

Examples might include:

🔸 pricing optimization
🔸 campaign targeting
🔸 inventory planning
🔸 product performance analysis

Decisions define the product’s purpose.

2️⃣ 𝗜𝗱𝗲𝗻𝘁𝗶𝗳𝘆 𝗿𝗲𝗮𝗹 𝘂𝘀𝗲𝗿𝘀

Every Data Product should have clearly defined consumers.

These might include:

🔸 analytics teams
🔸 operational systems
🔸 machine learning models
🔸 business applications

If nobody depends on the product, adoption will never happen.

3️⃣ 𝗗𝗲𝘀𝗶𝗴𝗻 𝘁𝗵𝗲 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝗶𝗻𝘁𝗲𝗿𝗳𝗮𝗰𝗲

Data Products are consumed in different ways.

Common interfaces include:

🔸 curated datasets
🔸 APIs
🔸 semantic layers
🔸 event streams

The interface determines how easily teams can integrate the product into their workflows.

4️⃣ 𝗗𝗲𝗳𝗶𝗻𝗲 𝗮 𝗱𝗮𝘁𝗮 𝗰𝗼𝗻𝘁𝗿𝗮𝗰𝘁

This is where many initiatives fail.

A good Data Product clearly defines:

🔸 schema and definitions
🔸 refresh frequency
🔸 reliability expectations
🔸 versioning rules

Consumers need stability as much as accuracy.

Krzysztof Wiśniewski, VP Data Engineering at Striped Giraffe:

“When Data Products are designed with users, interfaces, and contracts in mind, they stop being passive data assets. They become infrastructure for decision-making.”

🤔 How does your organization design its Data Products?
Let us know in the comments.

Striped Giraffe Innovation & Strategy GmbH

10/06/2026

09/06/2026

09/06/2026

08/06/2026

28/05/2026

27/05/2026

21/05/2026

20/05/2026

18/05/2026

13/05/2026

Adresse

Telefon

Webseite

Benachrichtigungen

Service Kontaktieren

Verknüpfungen

Teilen

Kategorie