Product Manager & UX Consultant

Alumnet Website

Project Type

  • Social networking startup

  • Redesign

Skills used

  • Product strategy

  • Roadmapping

  • UX research

  • User interviews

  • Analysis

  • Wireframing

  • Prototyping

Tools used

  • JIRA

  • Google Analytics

  • Sketch

  • InVision

  • Typeform

  • Hotjar

  • Unbounce


January 2018 - September 2018 (9 Months)

Alumnet is an early stage startup building a brand-new alumni engagement platform for schools in the UK

At the start of the year my focus moved from the iOS app onto the Alumnet website, working closely with the CEO to create the overall product strategy for 2018. I opted to set quarterly objectives and key results (OKRs) alongside creating a more traditional product roadmap to detail the improvements we needed to make to the existing MVP. I made sure to involve the whole team with the process of setting of the OKRs which kept everyone aligned, motivated and invested in their success.

Below are some examples of the feature development and improvements that were made.


One of Alumnet’s riskiest assumptions was that older alumni would be engaged in offering help and advice to recent graduates. Whilst the motivation for graduates wanting to leverage their alumni network for advice and job opportunities had been established, it was less clear that the older alumni would be motivated to offer this help.


By conducting interviews and surveying existing users, I would be able to understand their motivations for using Alumnet and quantify their willingness to offer advice.


The quantitative and qualitative results demonstrated a clear willingness to help from the older alumni, especially with regards to career advice. In addition to this, career advice was again qualified as a user need and motivation to use the network by the recent graduates. These results allowed us to better understand our user groups and establish early personas that would help inform our efforts to build a community, matching each group’s motivations with the other’s willingness - and later ability  - to offer help.

An example of one of the surveys results demonstrating users' willingness to offer career advice.

An example of one of the surveys results demonstrating users' willingness to offer career advice.


The registration form for the existing MVP had a poor conversion rate, leading to wasted marketing efforts.


Using data from Google Analytics, user recordings from Hotjar and user interviews, the following quick wins could be implemented to increase the conversion rate of the registration form:

  • Reducing the number of fields on the registration form (the final form was negotiated with all stakeholders)

  • Adding a button to allow users to prefill the form with data from their LinkedIn accounts

  • Creating a MVP version of a dynamic registration page that was specific to each school, pre-filling the school input field with the school name (e.g Harrow School) and including the alumni nickname of the school in the header (e.g Old Harrovians).


  • The bounce rate on the registration page fell by 12%

  • 23% of registrations in the following sprint used the pre-fill with LinkedIn button (the average now is at 26%)

  • The registration conversion rate increased by 14%

  • The dynamic registration landing pages (school specific) converted 21% better than the original form


Following the results of the dynamic registration pages that were school specific, I further personalised the acquisition journey by creating dynamic landing pages for each school, the CTAs of which would lead to the dynamic registration pages we had created.

You can see an example of one of the landing pages below:

Iteration Results

Visits following this more personalised registration journey convert a further 18% higher than the previous iteration.

This meant a 39% total uplift in registration conversion.


The user onboarding journey for the MVP had several issues. The drop out rate was high, it wasn’t giving us enough useful data about the users, and as a single journey, it didn’t differentiate between our known user groups often leading to the users being asked for irrelevant data.


  • Following the findings of the previous user research, we could now group the user’s motivation for using Alumnet into at least 1 of 4 categories

  • Adding a clear explanation as to why we were requesting the data at each step of the onboarding journey and detailing how it would benefit the user and their experience, would help lower the drop off rate in the funnel

  • Segmenting the user's journey to be conditional on their career stage, would prevent irrelevant questions being asked and also allow us to ask more specific questions around their personal circumstance

  • As users had proved their willingness to help other alumni in the previous surveys, adding a step to the onboarding that would allow them to signal the way in which they were willing to help would help add immediate value to the network


This redesigned onboarding journey increased conversion by 47%, whilst obtaining better and more useful data that was used to increase engagement across the platform

Click through the carousel below to see an example of one of the dynamic onboarding journeys.

As stakeholders wanted information that was collected in this previous journey but not included in the redesign, I negotiated for the data to be obtained at a later date, by designing prompts that would direct users to their profile, where they could add this data (house, sports played etc.)


The events section of the website provided the potential to be a very useful tool for acquisition. However, the MVP didn’t allow invites to people who hadn’t already joined the network.


A redesigned event page which included a unique referral link would aid conversion and help drive acquisition.

On receiving the referral link, invitee would be taken to a landing page that would show limited details of the event, the details of the person that had invited them and the option to login/register to see the rest of the details and RSVP.

You can follow the journey in a prototype by clicking here or by clicking the image of the event page below.