For years, many organisations have relied on a traditional analytics-as-a-service model: the business submits a request, it enters an analytics queue, and eventually an insight or report comes back. It worked for a time. But today, it’s slowing organisations down and limiting their ability to act with precision, speed and confidence.
In a recent conversation between Dufrain and Tesco Bank, the Tesco Bank team explained exactly why the old model no longer meets the needs of a modern financial services organisation and what needs to change instead.
The problem with the traditional analytics model
As Tesco Bank’s Head of Data Science, Gwil Morrison, described, analytics-as-a-service tends to create long lead times and bottlenecks. Requests sit in a prioritisation funnel while analytics teams juggle competing demands. For business leaders, this often means waiting on insights that are critical to commercial decisions.
Key challenges highlighted include:
- Slow time-to-insight as requests wait in a queue
- Misaligned priorities between analytics teams and business stakeholders
- Frustration and inefficiency, especially when decisions depend on timely analysis
This becomes particularly limiting in fast-moving markets, such as pricing, lending or customer behaviour modelling.
A “factory model” that holds teams back
Tesco Bank’s Head of Decision Science, Laura Castro, expanded on the constraints of the old approach. Teams often operate like a factory: producing repetitive outputs, monthly reports and one-off datasets with limited reusability.
Under this model:
- Stakeholders ask for X, and analytics teams deliver X every month, in the same way
- There’s little time to improve processes or automate recurring work
- Valuable talent ends up “cranking data” instead of solving meaningful problems
- Reproducibility and scalability become afterthoughts
The result? Teams stay trapped in reactive cycles instead of creating long-term value.
The impact on commercial performance
For Tesco Bank’s pricing function, the old model simply couldn’t keep pace with business needs. Chris Russell, Head of Pricing, summarised the issue clearly: his team operates in a live market where speed matters. Waiting in an analytics queue slowed decision-making and limited the depth of analysis his team could perform.
Even when insights arrived, they often raised new questions which triggered more requests and further delays.
Why a product mindset is the answer
The discussion made one point overwhelmingly clear: organisations need a model that delivers speed, depth, self-serve capability and repeatability.
This is where analytics-as-a-product comes in.
Instead of fulfilling one-off requests, analytics and business teams co-create reusable, scalable analytical products, toolkits, automated pipelines, models and front-end interfaces that empower teams to explore their own data with confidence.
This shift transforms:
- Time-to-insight
- Operational efficiency
- Cross-functional collaboration
- The quality of decisions made across the organisation
Toward a more agile analytics future
Analytics-as-a-service hasn’t failed because teams lack capability it has failed because organisations have outgrown it. Today’s financial services environment requires:
- Real-time understanding
- Faster experimentation
- Greater commercial responsiveness
- Reusable analytics assets
- Data products that evolve with business needs
The traditional model wasn’t built for this. Analytics-as-a-product is.
In Blog 2, we explore exactly how Tesco Bank made the shift – and the lessons leaders can apply as they begin their own journey. Stay tuned for more by following us on LinkedIn.
Watch the full discussion here.