UX STRATEGY CASE STUDY
From 8,000 to 11,000,000+ users: Product-market fit the human-centred way
Research & product strategy for the leading collaborative, online LaTeX editor trusted by 19 million users at research institutions and businesses worldwide for scientific writing & publishing
Key skills
/ ux strategy
to shape human-centred product prioritisation (Kano modelling, HEART metrics)
/ user research
strategy, planning and hands-on support (Qualtrics, mixed methods)
/ design leadership & product strategy
Sector
B2C, SaaS for researchers and technical writers globally
Timeline
Multiple projects from initial strategic investment in 2014 until I left Digital Science in 2021
Client
Overleaf - in 2014 Digital Science invested in WriteLatex (later rebranded to Overleaf) and as Head of UX, I supported the founders with strategic design & research priorities.
Challenges
As Head of UX at Digital Science, I was always working with a range of companies and product teams at the same time. My support of Overleaf was thus periodic and mostly of advisory nature, with bursts of hands-on support when most needed, and as highlighted in the engagements below.
Goals
Support the founders from initial investment through central UX support in a strategic manner to find product market fit, grow the product, and the business overall into a successful Digital Science investment.
Human-centred product strategy
in 3 parts
The backstory
In 2014, Digital Science invested in Overleaf (WriteLaTeX at the time).
As with many of our investments, I would sit down with the founders to explore ways to support them early on. Their product, Overleaf, offers real-time collaboration and WYSIWYG editing, making professional typesetting accessible to scientists and technical writers. The platform streamlines the entire scientific process, from idea to publication, aiming to make research faster, more open, and accessible.
For a seed stage investment, Overleaf had a healthy number of thousands of active users. This made the product an ideal candidate for a new strategic user engagement program I was creating alongside a similar initiative at our sister business - the Nature Publishing Group.
Part 1:
Seed stage
Using Qualtrics, I worked with the founders to create an effective user survey utilising a mix of standard usability questions but more importantly using Kano Modelling to quantitively assess the desirability of existing and future functionality that the founders were contemplating based on anecdotal evidence.
As usual for most product teams, the expectation is to add new features, to address complaints from users, requests, or to open up new customer segments. The results from the survey came out loud and clear that by far the most important effort would not be a new feature, but rather to improve an existing - seemingly fine working - feature. As we are talking about a very early stage product with a quite limited feature set, that result was somewhat of a surprise.
Shaping product strategy
by measuring desirability of future efforts
The founders' academic background in mathematics (PhDs) likely contributed to their receptiveness to my arguments and the clear data. I believe that this early collaboration, following Digital Science's seed-stage investment in Overleaf, was instrumental in shaping the company's ongoing commitment to a human-centric approach.
The team went ahead and spent the time to rebuild the PDF renderer. This was a resounding success and very well received by the users and laid the foundation for future product improvements. It also build a lot of trust and commitment in thorough, methodological engagement of the user base which laid the foundations for my future engagements with Overleaf described below.
A core part of Overleaf is the PDF renderer that takes the LaTeX code and translates it into the type-set publishable artefact that every researcher is aiming for.
The survey feedback indicated that the renderer was too slow, which significantly hindered the experience of writing and reviewing code, text, and formulas for the paper in the rendered PDF. And more importantly, the Kano modelling highlighted that improving the PDF rendering was the single most important improvement to users.
This presented a dilemma: improving the renderer's performance would require a complete rewrite, taking approximately six months. The founders had to choose between dedicating this time to enhance an existing feature or using it to develop new features.
Thanks to the rigorous methodology we employed, particularly the Kano model, and my persuasive communication, the founders made an informed decision to prioritize what users identified as their top concern.
A tough choice
To rebuild or not to rebuild
I recognized both the need for a more relevant approach and the willingness of our start-ups to embrace such a shift.
To better support our portfolio businesses, I devised a strategy to identify more appropriate metrics that could be implemented alongside the top-down mandated content-centric ones.
My efforts culminated in the successful implementation of these tailored metrics across Digital Science, with Overleaf emerging as a standout example of the initiative's effectiveness.
Under the ownership of Macmillan Publishers, Digital Science faced a significant challenge. A new metrics-driven strategy was introduced that catered for the larger sibling business, Nature Publishing. The new shift primarily focused on content-related metrics suitable to a publishing business but largely inappropriate or even counter productive to Digital Science's suite of productivity-oriented software tools.
Recognizing the disconnect between these metrics and the needs of Digital Science's businesses, many of the startup founders began pushing back. While there was an understanding that these metrics had to be reported due to corporate expectations, they were not seen as valuable for driving business success within Digital Science.
Phase 2:
Growth stage
After extensive research, I selected the the HEART framework developed by Rodden et al. at Google. This framework focuses on user-centric metrics and much better aligns with the needs of our portfolio business' needs - allowing us to measure what truly mattered.
I took my proposal to our CFO and secured buy-in to back this new strategy. I then worked closely with each of our portfolio businesses - founders, CTOs, or product leadership, depending on the nature and maturity of the business.
The playbook worked like this:
-
Together, we agree together what stage the business is in. This informs which category of metrics is most relevant.
-
Together, we then work on defining the desirable goals and translating them into signals and metrics that are both meaningful and measurable.
Shaping product evolution
with metrics that matter
By adopting and tailoring the HEART framework metrics to fit the unique needs of our startups, we created a more effective measurement system. Overleaf's success in embracing these metrics highlights the impact of aligning business metrics with the specific needs and stages of each company.
My contribution was instrumental in shifting the focus from generic, content-driven metrics to those that truly mattered, fostering a more effective measurement system that supported the success of our businesses at large and Overleaf in particular.
I worked directly with Overleaf's CTO to jointly select the goals and signals most appropriate for the current stage of the business. User priorities and feasibility to implement the underpinning metrics fed into the decision process. The most suitable metrics were then implemented by Overleaf, and reported back to the parent company on a monthly basis.
Out of our various portfolio companies, Overleaf stood out as the most committed participant in this company wide metrics strategy. They not only took the implementation seriously but also fully bought into continuously utilising the metric to inform a more user-centric approach in driving the success of the business.
Collaborating with technical leadership
to implement metrics that drive human-centred product strategy
In 2017, Digital Science bought the Overleaf competitor ShareLaTeX. The driving aim behind the acquisition was to bring the best of both editors together in a way that benefits everyone.
At the time, their combined community comprised approx. two million authors at thousands of institutions and enterprises around the world, ranging from students writing up their first group projects to professional researchers collaborating on their papers, grant proposals and more.
Both teams were being merged together with the goal to select and/or create one combined product going forward.
In my role as Head of UX for Digital Science I worked with Overleaf's leadership team (primarily with the CTO and the UX lead) on creating a user-centred, research-led strategy to inform the merging process.
Phase 3:
Market dominance
The methodology and execution of our approach won best case study award at ACM CHI 2021 for our case study on "Merging SaaS Products in A User-Centered Way - A Case Study of Overleaf and ShareLaTeX" (1).
ACM CHI is generally considered the most prestigious conference in the field of human–computer interaction and is one of the top-ranked conferences in computer science (2).
[1] https://programs.sigchi.org/chi/2021/program/content/56988
[2] https://en.wikipedia.org/wiki/Conference_on_Human_Factors_in_Computing_Systems
The mixed methods approach
A particular validated strength of our approach was the combination of large scale surveys pre- and post merger (>7,000 responses) and multiple targeted private beta releases targeting qualitative feedback.
The strategy
The devised user-centred strategy to merging the two products focused on
-
clear transparent communication,
-
robust qualitative and quantitative user research and testing,
-
iterative, gradual feature releases
The challenge
As is often the case with software mergers, diverging sets of functionality mean it is easy to alienate users and often hard to agree on what functionality and type of user experience to prioritise in a new combined offering.
An (award winning) mixed methods research strategy
to inform product priorities for the merger
Outcomes
Overleaf is a startup and social enterprise that builds modern collaborative authoring tools to help make science and research faster, more open and more transparent.
Overleaf’s market-leading collaboration technology is now in use by over 17 million researchers, students, and teachers in institutions, labs, and industry worldwide.
Overleaf is the world’s market leader in this collaboration technology, with more than 10 million users worldwide. The service assists academics, students and organizations to effectively and efficiently collaborate, publish and share their work, helping to make science and research faster, more accessible and more transparent.
-
189 countries represented
-
More than 6800 universities with students, faculty and staff using Overleaf
-
More than 2000 companies and other research institutions
-
11 million collaborations
-
87 million projects
Overleaf has won many awards
Sebastian Peck
(ex CFO @ Digital Science)
Sven is a passionate and indefatigable advocate of UX design excellence, spreading best practice with a view of making key UX tasks such as user research, prototyping and metrics-driven design a priority. He continuously identifies areas for improvement and ensures successful implementation through effective stakeholder management.
Mustafa Kurtuldu
(ex UX Designer @ Digital Science)
Sven is one of the most analytical and principled designers I have ever met. His knowledge on UX research, such as use of the kano model to applied design prototyping, graphic and illustration work make him a rarity in our industry. As Head of UX at Digital Science he was balanced and fair, willing to listen and active in leading design best practice, establishing standards across all of Digital Science portfolio companies. I feel privileged to have worked with him.
Steve Scott
(Director Portfolio Development @ Digital Science)
Working at the intersection of business, product and engineering is challenging at the best of times - in a startup even more so - but Sven's diplomatic skills helped him navigate through any friction.