Turning unstructured data into insights: interview with Squirro
Artificial Intelligence is hot on the agenda this year with every “Top Fintech Trends of 2018” list not complete without mention of the incoming robots that are going to be invading our offices to take our jobs with their timesaving algorithms.
We got to sit down with the kind folks over at Squirro who have brought together Artificial Intelligence and Machine Learning to provide Augmented Intelligence Solutions that give bankers (and generally us humans) more time to focus on the things that really matter.
What challenge did you set out solve?
Most organisations in financial services suffer from data overload and are challenged in finding the nuggets of gold in that vast sea of data. This is particularly true in wholesale banking, where emails, news feeds, analyst reports and internal data all form the rich tapestry of client engagement and deal origination. Hours of a banker’s valuable time can be spent gathering the data, analysing it and acting on it – this is what Squirro can address and automate; by turning that unstructured, but important, data into meaningful insights.
Our four founding partners had solved similar problems in a previous start-up, whereby they adopted a mathematical approach to address the challenges of analysing unstructured data. They used their experience to come together and build a solution that could mathematically match disorganised data to the business requirement as opposed to manually trawling through the data and thus Squirro was born.
What’s your core technology about and what makes you competitive?
Squirro provides Augmented Intelligence solutions. Its unique technology marries Artificial Intelligence, Machine Learning and predictive analytics, empowering organizations to transform enterprise data into AI-driven insights.
One of Squirro’s customers describes us as an antidote to complexity, so that’s a good place to start. Squirro’s route to value is typically measured in weeks, rather than months, with an agile and iterative approach to meeting and refining the desired business outcomes.
From a technology perspective, Squirro’s probabilistic approach to finding those nuggets of information needed by the business users and Squirro’s patented Smart Filter technology are unique in the cognitive insights’ space.
The core Squirro team have been together through a number of search-related start-ups, with their insights enabling development of this unique capability. We also dedicate more than 50% of our resources to R&D.
Who benefits from your systems?
Squirro is suited to all vertical industries with a focus on Financial Solutions, including Corporate & Institutional Banking, Investment Banking, Real Estate Finance and Institutional Asset Management, Corporate Insurance and Manufacturing.
You can see for yourself what Squirro’s customers say about them below.
Do you foresee a “data drought” in the EU with GDPR and will this affect you?
GDPR will not necessarily bring about a data drought if the process is managed proactively.
As an example, one UK charity proactively approached their donors over a year before GDPR was due to come into effect and actually increased the number of contacts for whom they had consent by outlining the positives aspects of keeping in touch.
It is understandable that in B2C relationships, GDPR will have an initial impact on marketing capabilities. In our world, our customers to which we provide enrichment are typically corporate customers. Data is often brought in often from the public domain or held internally with a contractual basis for retention. Because of this we haven’t seen a significant impact on available data with GDPR.
Will AI replace human intelligence?
AI and machine learning are certainly topical at the moment but we think the focus should be on not just the technology but the desired business outcomes. Squirro delivers Augmented Intelligence solutions that address the time-consuming processes of gathering unstructured data to support client, market or opportunity analysis and then providing what-to-do-next action recommendations and business insights.
So to answer your question, we won’t replace human intelligence – we address the lower value, high volume workload to free up human time for more valuable tasks.
Arguably, AI is the most important ‘disrupter’ for banks today, so at Squirro we were very interested in the uptake so recently commissioned a survey and found the following insights.
The use of artificial intelligence (AI) and machine learning (ML) in financial services (FS) is on the rise, with 83% of banks having evaluated AI & ML solutions
67% have actively deployed them.
There is a lack of understanding around AI & ML, with 83% of respondents still unaware of how to apply the technology to solve business problems.
Using AI and machine learning to source new leads and opportunities is key to bankers, with 87% of respondents saying that it would be highly impactful if an AI engine could spot relevant events that led to engaging with a client and closing a deal.
Bankers recognise that AI driven insights will have a tremendous impact when it comes to anticipating market events to stay ahead of the competition.
Squirro’s ability to ingest unstructured data from both internal and external sources and then to apply AI and ML to the curation and analysis of that data enables a wide range of possible solutions, but as our recent study has validated, our focus on providing support towards suggesting original opportunities is key across the banking domain.
In summary the primary risk with AI and ML is that enthusiasm for the technology may deflect from what should be the focus on the business outcomes. AI and ML are only meaningful when they are embedded into a solution that delivers business outcomes that align with the organisation’s strategic direction.
Having said that, AI and ML are prolific components in many business solutions today, so clients need to be aware of the technology as it evolves to ensure they are leveraging it successfully to achieve the productivity gains and insights that are enabled by it.