Our Service Areas
Three services,
each with a clear scope
Tessera works across natural language analytics, recommendation systems, and data pipeline engineering. Each engagement is defined upfront and delivered within an agreed timeframe.
Back to Home— Our Methodology
How we approach each engagement
Every Tessera engagement follows the same structure: a scoping conversation to understand what is needed, a written scope document that defines deliverables and success criteria, delivery with regular progress updates, and a structured handover session at the close.
The pace of each engagement is set by the project type. Data pipeline work tends to be methodical and linear. NLP analytics involves iteration on taxonomy and modelling. Recommendation systems require a period of data analysis before architecture decisions can be made. Each timeline reflects these realities.
Scoping conversation
We discuss your situation, data environment, and what you are hoping to achieve. No commitment required at this stage.
Scope document agreed
Deliverables, timeline, success criteria, and data handling terms are confirmed in writing before any work begins.
Delivery with updates
Work proceeds with regular progress communication. Any issues or required adjustments are raised early.
Handover & documentation
The engagement closes with written documentation and a session where your team learns to operate what was built.
— Service 01
Natural Language Analytics
RM 2,700 · 8-week engagement
This service focuses on extracting structured insight from text-based data: customer feedback, support tickets, survey responses, and similar written inputs. Many organisations collect this data routinely but review it manually, or not at all. The volume makes consistent review difficult and leaves patterns unidentified.
Tessera's NLP analytics work covers the full process — from designing the taxonomy used to categorise text, through sentiment modelling, to dashboard creation for ongoing visibility into what your written data is telling you.
What is included
- Taxonomy design appropriate to your data categories
- Sentiment and intent modelling
- Dashboard creation for ongoing insight review
- Full documentation and handover session
- Configuration and reporting within 8 weeks
Suitable for teams that
- Collect substantial written feedback or survey data
- Want to identify themes and patterns without manual review
- Need a structured way to track sentiment over time
Process overview
Week 1–2: Data review and taxonomy design
Week 3–5: Sentiment model development and testing
Week 6–7: Dashboard configuration and refinement
Week 8: Final review, documentation, handover
— Service 02
Recommendation Engine Design
RM 3,600 · 10-week engagement + 2-week optimisation
A recommendation engine suggests relevant options to users based on their observed behaviour or stated preferences. This applies to product recommendations in e-commerce, content surfacing in media platforms, and knowledge management systems for internal teams.
The engagement covers user behaviour analysis to understand what patterns exist in your data, architecture and implementation of the filtering approach that fits those patterns, and a testing framework so you can measure the system's impact after it is live.
What is included
- User behaviour analysis from your existing data
- Collaborative filtering design and implementation
- A/B testing framework setup
- 2-week optimisation window post-deployment
- Full documentation and handover session
Suitable for organisations that
- Have existing user behaviour or interaction data
- Want to surface relevant options without manual curation
- Operate an e-commerce, content, or knowledge platform
Process overview
Week 1–3: Behaviour analysis and architecture design
Week 4–7: Filtering system build and internal testing
Week 8–9: A/B framework setup and deployment prep
Week 10–12: Deployment, optimisation, handover
— Service 03
Data Pipeline Optimisation
RM 1,500 · ~6-week engagement
When data infrastructure creates bottlenecks — slow processing, inconsistent outputs, difficulty feeding analytical systems — the underlying cause is usually in the pipeline design rather than the tools being used. This service reviews and restructures existing data flows to address those bottlenecks directly.
The work covers schema normalisation to create consistent data structures, ETL process redesign to reduce latency and processing overhead, and monitoring setup so your team can detect issues before they affect downstream systems. This is also the recommended starting point for organisations planning to add AI capabilities — a well-structured pipeline makes every subsequent step more straightforward.
What is included
- Review of existing data flows and bottleneck identification
- Schema normalisation for consistent data structures
- ETL process redesign and latency reduction
- Monitoring setup for ongoing health tracking
- Full documentation and handover session
Suitable for teams whose
- Data infrastructure creates delays in analytical reporting
- Current pipelines produce inconsistent or unreliable outputs
- AI system plans require a more reliable data foundation
Process overview
Week 1–2: Pipeline review and bottleneck mapping
Week 3–4: Schema normalisation and ETL redesign
Week 5: Monitoring configuration and testing
Week 6: Final validation, documentation, handover
— Comparison
Which service is right for you?
Use this overview to understand what each engagement covers and which situation it best fits. More than one may be relevant — they are designed to work together.
| Feature | NL Analytics | Recommendation Engine | Data Pipeline |
|---|---|---|---|
| Price (RM) | 2,700 | 3,600 | 1,500 |
| Duration | 8 weeks | 10 + 2 weeks | ~6 weeks |
| Text/feedback data focus | |||
| User behaviour analysis | |||
| Data infrastructure work | |||
| Dashboard or monitoring output | |||
| A/B testing framework included | |||
| Good foundation for future AI work |
Not sure where to start? If your organisation is new to structured AI work, Data Pipeline Optimisation is often the most practical first step — it addresses foundational issues that affect everything that follows and is the most accessible entry point by cost and scope.
— Standards
Shared across all engagements
Privacy & Data Handling
Processing boundaries are documented before work begins. Client data is not retained after handover. All work is conducted in compliance with Malaysia's Personal Data Protection Act.
Testing Before Handover
All deliverables are validated against agreed success criteria before the engagement closes. Adjustments are made within scope if testing reveals shortfalls.
Written Documentation
System architecture, configuration, and operational guidance are provided in writing at the close of every engagement — sufficient for your team to manage the system independently.
Modular Outputs
Systems are built with future extension in mind. Where possible, outputs from one engagement can feed into or complement subsequent work — without requiring a rebuild.
Clear Communication
Progress is communicated at a regular cadence throughout each engagement. Issues and scope considerations are raised early, not buffered to project end.
Knowledge Transfer
A structured handover session is included in every engagement. Your team finishes the project able to operate and maintain what was built without ongoing Tessera involvement.
— Pricing
Transparent, fixed pricing
Data Pipeline
RM 1,500
~6 weeks
- Pipeline review & restructure
- ETL redesign
- Monitoring setup
- Documentation & handover
NL Analytics
RM 2,700
8 weeks
- Taxonomy design
- Sentiment modelling
- Dashboard creation
- Documentation & handover
Recommendation Engine
RM 3,600
10 + 2 weeks
- Behaviour analysis
- Filtering system build
- A/B testing framework
- Documentation & handover
Have a project in mind?
Get in touch to discuss scope and approach. A conversation is a good first step before any decision is made.
Contact Tessera