Re-envisioning how analysts at Venture Capital firms can efficiently discover companies of interest on Synaptic's web app.
Team
Co-founder
Anurag Abbott
PM
Harsha
Engineers
Nikhilesh, Sameer, Vaibhav
Designers
Rupak Mishra, Jahnavi Kolakaluri
My role
Conducted in an in-depth comparative study on advanced filters, ideated and proposed design solutions to the team, created user flows, visual design, and components for the filtering interface. Collaborated with engineers to assess the functioning of the filters.
Synaptic is a Series B start up that offers Ventures Capital firms alternative data insights such as - reviews that customers leave about a company’s products, or the user traffic to a company's website.
Analysts at VC firms then use this data to assess whether they want to invest/fund in these companies or not.
What is Screener?
Screener is a no-code web platform by Synaptic that let’s analysts filter through company data and view companies based on their desired criteria such as - company location or funding amount.
problem space
Analysts want to build Screeners that are more customizable. Such as those that show companies:
Show me companies in Asia, with a Total Funding Amount > 10M or a Total Funding Amount between 1M to 10M and is a Series B company. With a Glassdoor Overall Rating (YOY%) > 50% given by full time employees.
But too-many dropdown filters and their limited complexity has led to increased time and decreased complexity when building a Screener.
This led to  workarounds  that led the Screener interface ineffective, highlighting the importance of this issue.
THEÂ SOLUTION
The common dropdown presents all conditions/fields the user wants to filter by. This makes it easier to access these conditions compared to the numerous dropdowns present across the screen, leading to confusion and increased time.
This leads to the filter (a condition) being constructed as a sentence with the field/condition + operator (is, is not) + value. Multiple of these can be stringed together. Here user asks for companies in the United States and with a Total Funding amount equal to 10M dollars.
Defined easy to use value dropdowns for text, number values, and location and more. This is vital as having faster selection and easy to use value drop down allows for faster construction of the filters.
Grouping filter feature allows a set of filters to be considered as one filter. This allows for more complexity and customizability (i.e. ability to ask multiple conditions at the same time.
Sub conditions give more granular control over the filters. This allows users to refine a condition, for example to ask Glassdoor overall rating this is given by only full time employees.
my impact
1. I was able to empower users AND businesses.
2. I played a pivotal role in the revamp of Screener. During my time at Synaptic I broke down a complex problem to construct a new, logically sound, and implementable filtering system.
3. I contributed to the new design system for filter components during a time when the design system was being revamped. I experimented heavily to achieve visual balance through my mockups. I used different colors for each filter to help distinguish filters on a functional level but also bring out visual personality. A mixture of lighter shades for the futuristic feel and bold text and outlines to give off a strong personality.
design process
Our users are asking for more specific and nuanced data, and the current dropdown filters, are just not accommodating for these.
identifying pain points
As an intern, and because of the specific nature of the work analysts do, I was not able to directly interact/ observe users using Screener. I was proactive in asking questions to our founders, designers and product managers, as they hand accumulated subject matter over time, and had previous interactions with users. Through cognitive walkthroughs I identified the following limitations on the current platform:
1. Inability to construct complex queries
2. Inconsistent use of drop down filters
3. Visual & Interaction Inconsistency
4. Lack of Scalability
Comparative study & Key takeaways
1. The Common Dropdown
To filter down to a particular dataset, the user needs to choose a condition or field to filter by. To make adding conditions more intuitive and repeatable, several of these platforms encase every filtering criterion in one large drop down. This eliminates the struggle for ‘searching’ for a criterion on the interface and allows easy access to all fields in one place.
2. General Filter Structure
While a trivial observation, I noticed that the filter structure itself follows a general pattern of field + operator + value. The fields also had several user facing data types: like numbers, words, and dates. Now, the operators matched the data types they were operating on. Numbers could be compared with operators like ‘less than’, or ‘between’.
3. Adding Filters with AND/OR
The use of AND/OR to stitch filters together to filter multiple conditions. Cruchbase uses a different layout, conveying the filtering as a mind map.
4. Grouping Filters
Grouping allows combining several smaller conditions under a single umbrella condition. This allows us to treat these conditions as one larger group, which is much easier to deal with. It is also very handy when the user wants to create complex queries without feeling the burden of dealing with several same-looking conditions with no hierarchy.
5. Sub Conditions
Sub conditions give more granular control over the filters.
initial ideation
Approach 1
Approach 1 starts with asking questions in natural language and then breaks it down into what metrics the analyst wants to tweak. The metrics can be further adjusted after the Smart Query is processed - as shown in Approach 2. Approach 1 was however an engineering challenge in the short term. . After discussion we decided to focus more on Approach 2 because it's suits user needs, and meets engineering requirements.