Data Science.

With years of experience in data strategy and engineering, our team is uniquely placed to assess your business strategy and create a bespoke data solution that fuels your growth.

Find the why behind the what.

1. Value Mining

We help you ask the right questions and identify growth opportunities hidden in your data.

2. NLP

Turn vast amount of unstructured text data into enriched and easily-queryable attributes and discover what people really think.

3. Customer Lifetime Value

Keep hold of those who will grow alongside you with early signals applied to campaigns, ads, products etc.

4. Churn Risk

Understand churn risk on an individual basis and. embed risk scores into product flows, CRMs & sales tools.

5. Engage & Convert

Propensity models can give you an indication of how likely prospects will engage with marketing and who is more likely to convert.

6. Spend Incrementality

Using complex algorithms we can help to model the incremental return on campaign spend and find that ceiling where you stop having impact.

Technology We Work With

No two clients are the same and so our experienced team have built up knowledge of a wide range of different platforms/interfaces.

Here are the key technologies we work with. Don’t see your tech here and want to know if we can help? Get in touch!

Our Approach

No matter how much data you have, there is nothing less predictable than human nature! While propensity models are a useful indicator, they will always be a guiding light rather than a lazer beam. But that doesn’t mean they can’t have a huge impact on your business. Real-world experiments are often impractical and costly and while ‘let’s try it and see’ may have worked at one time, why take the risk when you can produce actionable insights from existing data?

For those using propensity models, they will most often take the form of scores. For example, your marketing department might want to understand the likelihood that different groups will open an email. A propensity score would show them how different groups or would respond to emails at different times, and with enough data could even show you on an individual basis.

Natural Language Processing uses several techniques to understand and manipulate human languages. Through text vectorisation, Natural Language Processing transforms texts into something that machines can understand (vectors of numbers). These transformed parts are fed to the system to create a machine learning algorithm. 

The statistical analysis approach then allows the software program to associate a particular type of output with given types of inputs. These outputs then undergo training so that the system can identify patterns within texts and make informed predictions about unseen data.

While understanding sentiment at scale can save you time sifting through the internet, we would advise though that sentiment analysis is still very rudimentary and can’t detect sarcasm, and so should be taken with a pinch of salt. NLP will often struggle to understand context and will have a tendency to pick up wrong or inadequate data, resulting in an inefficient learning process.

Ready To Start Your Data Engine?

Want to find out what the possibilities are for turning your data into business drivers?

Not sure if your ideas for using data are technically feasible and will generate ROI?

Have a specific use case in mind for your data but don’t know how to get started?

Get in Touch!

 

We are the perfect blend of strategic thinking and technical implementation.

hello@173tech.com

Old Street, London