Inside Open Source #8
All things Open Source: Interesting reads, startup news, trivia and views from Europe and beyond
Welcome to #8 of Inside Open Source
Every now and then I write about interesting developments in the Open Source ecosystem and related topics that I come across. Check out this issue below and subscribe if you like what you read :)
Topics:
🔃The rise of Open Source SaaS alternatives
🤴“From pauper-to-prince: developers are now the buying center of purchasing tools”
🔢45 Statistics, Facts & Forecasts on Machine Learning
🔃The rise of open-source SaaS alternatives
In recent years, many open-source alternatives to big SaaS players became very popular and raised a lot of funding. Companies like AirByte, PostHog, n8n, Supabase, Sentry are great examples of open-source alternatives to products like Fivetran, MixPanel, Zapier, Firebase and Datadog that were adopted by thousands of companies. These companies have a combined valuation of billions. This rise has driven the adoption of open-source software by enterprises.
Rajko Radovanović from NEA (Venture Capital fund that invested in Databricks, Cloudflare, Plaid, DataRobot, Anyscale, Pulumi, Tableau, MongoDB, Elastic among many others) wrote a great article about The Rise of Open Source Challengers.
Besides the general popularity that Open Source has gained in recent years, Rajko outlines why open-source SaaS alternatives are so successful and explains the recent wave of companies in the space.
According to him, many SaaS platform players are moving closer to integrating developer personas as key stakeholders for their product catalog. They extend their product offering by allowing devs to build all kinds of customized components. Take Airtable with their custom workflow automation, visualizations, and API-enabled integrations. Other prominent examples of big SaaS players moving closer to developer communities are Slack, Notion, Figma etc.
So the opportunity for open-source alternatives for these SaaS products arises when a large part of the value creation of customers with the product is enabled by the developer-facing product components. Open-source will always be a differentiating factor for some customers when there is certain proximity to developers as key stakeholders within the organizations. When a company uses Airtable for a certain business process and the internal setup is heavily dependent on custom made workflows and API-integrations, then there is a strong likelihood that the open-source alternative (NocoDB) might be considered as an alternative because of the general advantages that an open-source solution has against closed-core software from a developer’s perspective.
Obviously, the addressable markets for tools like Airtable, Notion etc are extremely huge so there is enough space for tools with such a differentiation towards certain customer segments and user personas. Although I doubt that is the case for every product category out there. If the closed-core SaaS tool has market size limitations there might be a hard cap on the addressable market that the open-source alternatives are able to target. So from my perspective, it’s not as simple as picking a SaaS unicorn with some developer adjacency and building a unicorn open-source alternative to it. What a shame, that would have been nice.
Many companies raised shiny funding rounds and some companies manifested themselves in the market but there has not been a company in this segment yet that reached the level of scale of the “original” SaaS comparable. Let’s see how that develops in the next years! I am sure some companies have the right ingredients to really take off big time.
If you want to check out a comprehensive list of open-source alternatives you should take a look at https://www.opensourcealternative.to/
Shoutout to Jonathan Reimer from crowd.dev (open-source developer community management software) for this nice collection.
🤴“From pauper-to-prince: developers are now the buying centre of purchasing tools”
In the podcast “Selling to Developers & Open Source Business Models” from a16z’s Peter Levin said the following:
One of the most notable transformations over the past 5 years has been the pauper-to-prince of the developer as a buying center within a company. […] As software infiltrates every part of our economy, [developers are now] the lead innovators and they’re the lead buyers in companies. What we see are many of our startup companies now deliberately selling to developers as the first wedge point into an organisation. [Developers] all very much have opinions and buying potential.
Slashdata (a developer community analytics company) took this statement and tried to get some hard data on it. Check out their findings below:
My highlights:
More than a third of developers (incl 38% of front-line developers) acquire tools for their personal use
Almost all developers with a leadership function are somehow involved in purchase decisions
Between half and two-thirds of developers are in a position to make recommendations or influence decisions. In particular, team leaders (i.e. senior developers) are big influencers (68%)
Up to a third of mid-level developer leaders (team leaders and product managers) are in charge of writing specs for tools (32% and 31%, respectively), or even get the final word on which tool to adopt (30% and 27%)
Only for budget and expense approvals, the power still lies elsewhere in the organization. Even on the CTO/CIO level, only 34% and 38% have budget and expense control
This was a survey from 2017 - imagine the transition within the last years. I haven’t found a recent report but if you come across something pls send it over!
🔢45 Statistics, Facts & Forecasts on Machine Learning
I recently came across this super nice collection of facts and figures from the Machine Learning space: https://research.aimultiple.com/ml-stats/
My highlights:
The value of global machine learning market was $8 billion in 2019 and is likely to reach USD 117 billion by the end of 2027 at a CAGR of 39%
46% of respondents have deployed ML in multiple areas and it is core to the business
10% of respondents are experimenting and investing in infrastructure and people
North America (80%) leads in ML adoption, and it is followed by Asia (37%) and Europe (29%)
Budgets for ML programs are growing most often by 25%, and the banking, manufacturing, and IT industries have seen the largest budget growth this year.
12.5% of employee time is lost in data collection. That’s five hours a week in a 40-hour workweek
40% of respondents indicated their organizations check for model fairness and bias. This ratio increases to one in two (54%) when only companies with extensive machine learning experience are included
Among companies with extensive experience in machine learning, only one in two (53%) companies check for data privacy implications in machine learning projects. The ratio drops to 43% when all companies are included
According to Refinitiv survey (2019), top challenges of machine learning adoption are
Poor data quality (43%)
Lack of data availability (38%)
Finding data science talent (33%)
According to Algorithmia survey (2020), top challenges of machine learning adoption are:
Scaling models (43%)
Versioning and reproducibility of models (41%)
Getting organizational alignment and senior buy-in for ML initiatives (34%)