Why Embedding Beats Outsourcing for Edtech Engineering
...and more on K12 cybersecurity initiatives, new AI tools from Chegg, Quizlet, and Flatiron, edtech industry funding updates, AI literacy, and more!
This week we have a guest article written by Sha-Mayn Teh (ex-Googler & former Engineering Director at TeachersPayTeachers) and Yu Chen (ex-Google sales and product leader), the co-founders of Potluck Tech. Potluck Tech offers fractional CTO/CPO services in combination with outsourced end-to-end development.
This article draws from the authors’ extensive experience building software, working with dev shops, and angel investing in startups.
Embed, Don't Outsource: Building Effective Development Partnerships in Edtech
In the current era of AI-fueled “micro-companies” and lean technical teams, Edtech companies are increasingly turning to outsourced software development companies or agencies to achieve the growth they need. These international companies are also known as “nearshore” (if in the same time zone) or “offshore” (in a different time zone) development firms, or more colloquially, “dev shops.”’
We don't see outsourced development going away anytime soon, but at the same time, leveraging dev shops effectively is much easier said than done. We’ve seen many instances where a company threw away an entire product their dev shop built, or where projects went way over budget and timeline.
In this article, we’ll advise Edtech companies on effective methods to identify the right dev shop for their needs, and what you can do to set up your collaboration for success.
What are Dev Shops, and Should I Be Working With One?
Dev shops are outsourced software development companies that may offer specific individuals for hire on a contract basis – which is known as “staff augmentation” – or even offer full development teams (including UI/UX, product managers, developers) that can build products from scratch. Dev shops may also specialize in specific industries (Edtech, health, fintech) or specific technology stacks (MERN, Python-Django, React Native, and other combinations of front and back end tools).
Many, if not most, tech companies employ dev shops at one point or another, yet working effectively with a dev shop is still confusing for many founders.
We have found that there are a few common situations that lead founders or engineering leads to consider working with a dev shop for the first time:
Their company has a non-technical founding team
Their company has a technical founder who needs more support
Their company needs additional expertise to complement an in-house engineering team
Each of these situations requires different needs from a dev shop, so let’s break them down:
Situation 1: Supporting a Non-Technical Founder
Non-technical founders can make significant progress with no-code and low-code tools, but they may still reach a point where they need technical help to build custom software, such as to provide deep personalization for their users or to meet ever-changing data privacy requirements.
For non-technical founders who hit a tech speed bump, one of their first instincts is to try to hire a technical lead to fill the gap. However, co-founder “dating” is a slow and painful process (much like regular dating!), and for early stage companies, it’s difficult to recruit a CTO or founding engineer without much traction or funding.
With DIY and full-time hires off the table, non-technical founders often begin evaluating outsourcing options. It makes sense: bringing on outsourced support is a great way to bridge expertise gaps until a company has the resources to hire in-house. It also gives them the ability to quickly flex up and down their technical capabilities in the future.
However, founders also need to tread carefully when they reach this milestone. It’s often difficult for non-technical founders to evaluate technical hires, and even harder for them to manage the cost, time and effort it’d take for engineers to build custom software. We’ll be talking more about how to improve your chances for success later on in this post.
Situation 2: Supporting a Technical Founder
Non-technical founders often think that if they have a technical co-founder, then all their product building problems are solved. However, this is not necessarily true! A technical co-founder is a precious resource, and in early-stage startups they’re almost always pulled in multiple directions. For example, when the technical founder needs to focus on fundraising, product velocity may suffer.
Leveraging outsourced teammates is one way to continue building product, and at a lower cost than hiring an engineer in-house. In fact, even when a technical co-founder is still writing code, we’ve seen startups complement the team with outsourced hires to accelerate velocity.
Situation 3: Supporting an In-House Engineering Team
Once companies have built in-house engineering teams, they still sometimes outsource development work! The company may need technical expertise in domains they don’t have in-house, or don’t want to staff full-time hires to do. For example, developer-facing product startups may employ contractors to maintain SDKs (software development kits) for less common languages or frameworks.
Alternatively, a startup may need to build quickly and can’t hire full time employees fast enough. For example, during COVID, many edtech companies saw massive surges in user traffic in a matter of days, and had to build more rapidly than usual. One instance of this was when Codecademy (which was already ten years old at the start of the pandemic) was able to quickly build a new feature with an outsourced team.
What Do I Need From a Dev Shop?
Edtech companies may have a pressing need for outsourced development at virtually any stage, from the early days when the founder is working alone all the way until maturity. Because this is the case, every Edtech founder or engineering manager should know how to work with outsourced development effectively.
And let’s be honest: evaluating outsourcing options is intimidating and time-consuming. Most people don’t have experience with outsourcing, and the cost of mistakes can be high.
The best move is to start by clarifying what you really need before jumping into an engagement, especially around the following criteria:
Scope and Duration: How Far Out Does Your Roadmap Go?
Do you have experience building product roadmaps? How clear is your roadmap and how far into the future does it go? The clearer your roadmap is, the more successful you’ll be in retaining an outsourced team on a project basis. You’ll need to spend valuable time upfront to create the product specifications, and continue to manage the build to avoid feature creep, but it’s possible to do this successfully. This is usually the most attractive option if your top priority is managing your budget and runway, and your product roadmap isn’t changing quickly.
That said, early stage technology start-ups often iterate far too quickly to be able to scope projects more than a few weeks out, and that’s where the time-based retainer model tends to be more successful. Depending on the startup’s budget and its velocity in responding to customer feedback, it should look into hiring outsourced engineers on a part-time basis (e.g., 20 hours a week, or 1.5 engineers spread across teams like Full-Stack, Mobile, and DevOps), and ramp the requirements up and down relatively quickly as needs change.
Location: How Remote-Friendly Are You?
How remote-friendly are you and your team? What time zone difference are you willing to work with? The following terms refer to where the developers are geographically located, and how close they are in terms of time-zone. Notice that the potential for quality work is there for all models, but they may have very different costs:
On-shore developers are generally in the US but are also the most expensive.
Near-shore is a relatively new term and means located in Canada, Mexico, or Latin America, often in the same time zone as the hiring company. The developers are slightly cheaper and usually in the same time zones, but it’s a more nascent model so there are fewer talent pools and agencies to tap into.
Off-shore is the traditional outsourcing model that almost everyone is familiar with. This is typically the most affordable option and features several high quality talent pools (e.g. Eastern Europe, Southeast Asia), but requires your company to be remote-first. Luckily, this has been the norm for many startups since COVID.
![Offshore vs Nearshore Development: Risks & Benefits [2023]](https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F7b0f46f1-7e04-4d16-b09c-022c5a7b68cf_1024x565.jpeg "Offshore vs Nearshore Development: Risks & Benefits [2023]")
Special Expertise for Edtech: How Are Your Regulation Chops? Can You Scale Seasonally?
Dev shops can provide specialized expertise that your in-house team might lack. This can be crucial for Edtech startups needing to navigate complex regulations for student data privacy (e.g. FERPA) or security controls (e.g. CIS). Find a dev shop that has experience with privacy requirements that will build privacy and security early on into your infrastructure design and operational processes. This can avoid headaches trying to retrofit these requirements later, and turn a challenge into a competitive advantage.
Infrastructure expertise may also be important. Edtech development is subject to unique seasonality effects. The summer is quiet, and that back to school sale might be a massive traffic spike your services will need to scale up for. If you’re developing for schools, your users likely have limited network bandwidth. Good backend and DevOps engineers will be able to scale your infrastructure to meet high quality service guarantees while keeping costs at a minimum.
Last But Not Least: Budget Budget Budget!
None of this matters if you can’t afford it. How do you work within your budget? Tradeoffs.
Project-based work is more predictable budget-wise but you’ll need to be very careful about managing scope creep, and your user-facing project is rarely “done” since you’ll need to iterate based on feedback. Off-shore work is cheaper than on-shore work but you’ll need to have intentional communication and documentation norms.
Be careful and upfront when discussing budget because cost and quality vary WIDELY: we’ve seen a 5x difference in pricing (!!) across outsourcing options.
Do your homework and look through freelance hiring platforms (Upwork, Toptal, terminal.io) quickly to establish a baseline understanding of the market. In the current market climate, you may even be able to snag some great talent at lower than their usual rate!
Keep in mind that you’ll likely pay more for a unique skill set, or if you want to hire on-shore. Also pay attention to the additional margins that agencies typically charge on top of developer rates – at the very least, they should be transparent.
How Do I Hire the Right Dev Shop?
Once you’re clear about your core needs, ask your network for referrals, including your contacts, LinkedIn, or other agencies like design or marketing companies.
Be as rigorous as if you were hiring a full-time employee.
Review the dev shop’s case studies, Clutch reviews and references, even if they are a referral. If you need help with technical screening, borrow time from experienced hiring managers who can assess technical skills, problem solving ability and sustainable engineering culture (testing, code reviews, etc).
It's important to remember that technical capability is not the only consideration: culture fit, growth mindset and a shared commitment to the product's success are equally important. As a tech-enabled startup, you’ll likely be pushing the boundaries of your collective knowledge and you’ll need to learn and grow together.
Note that incoming engineers are likely to disagree with how your prior technical team built out your product or software, so you’ll want to work with people who are open and adaptable. You might also want to start with a small trial project to test the working relationship before jumping all the way in.
How Can I Work Most Effectively With Dev Shops?
You’ve defined your needs, gotten referrals, and found the perfect dev shop. Congratulations!
Now the secret sauce to making the partnership successful lies in how you integrate them into your team. Here are some principles for building a healthy, high performing engineering culture with outsourced talent. Spoiler: these tips apply to your in-house teams, too!
Embed, Don’t Outsource
Don’t hand over a set of requirements and expect a finished product in return. In reality, tight collaboration is critical. Create cross-functional teams with both in-house and outsourced folks, give them clear goals, and do regular retrospectives to ensure things are going smoothly. This model is called staff augmentation, and it’s relatively common in the edtech ecosystem (e.g. MasterClass and Emeritus).
Embrace Iterative Development
The best software is developed continuously and iteratively in tandem with user feedback. Plan iterative cycles into your schedule and budget. Look for dev shops that follow agile development principles, like prioritizing frequent delivery and communication between business and development people. You should be seeing progress in your product every week!
Provide a Problem to Solve
The first step is agreeing on the problem (a simple but powerful principle to ensure alignment), then give the team ownership of the problem. In general, development teams perform best when they clearly understand the “why” and are empowered to solve it.
Prioritize the Virtual Water Cooler
Development teams thrive in a culture of learning and camaraderie, producing lots of ideas, creativity and solving problems proactively. Trust and strong project management play a vital role too. Create spaces and activities where people can interact and get to know each other’s culture and working styles.
Measure the Health of Your Process
In an early stage startup, you may not know what product to build, but you can focus on shipping fast with high quality. Track metrics for your team’s delivery, and ensure maintainability through testing code and monitoring your systems. If you’re a non-technical founder, consider getting external support from a fractional CTO or tech advisor to guide you on appropriate best practices for the stage of your company.
A Case Study: Dev Shops Gone Right
Here’s a story from personal experience: at TeachersPayTeachers (aka TpT), a dev shop, Wise Engineering, built and scaled the marketplace to millions of educators. I (Sha-Mayn) joined when the company was already 9 years old and at a pivotal moment of growth. I was tasked with building the in-house New York engineering team.
I was surprised at how much Wise was committed to the mission of the company. The TpT leadership team (including CEO Adam Freed and founder Paul Edelman) didn’t treat them like “devs for hire” and were very intentional about involving Wise in the company culture. Wise folks attended all-hands meetings, visited New York regularly, and everyone collaborating with Wise had the opportunity to visit their home town in western Ukraine. In short, they were considered part of the company.
As we staffed up in-house, we organized engineering into agile, autonomous pods (cross-functional development teams inspired by Spotify squads), and Wise engineers were embedded in almost all of them, except for one fully remote product pod. This turned out to be a productive model that aligned everyone to company goals and enabled us to tackle large technical and product initiatives as a team.
What was most notable in this case study?
It was important to define communication norms and ensure alignment with remote folks. This took a lot of training (this was B.C. - Before COVID). We moved all hallway chatter into Slack and documented all decisions in tickets and technical design documents. Wise’s CEO was an engineering lead involved in decision-making for the organization.
Since each team had off-shore engineers, we split on-call schedules into day/night shifts, allowing us to leverage time zone differences for easier 24x7 production support.
Everyone was included in engineering-wide activities e.g. hackathons, fixits and training. At the same time, Wise had a support system for their own employees that included training and mentorship.
Notably, most Wise engineers had been there for years, so they were well steeped in the TpT culture; in fact, having them on each pod helped to seed the culture and onboard newer engineers. Wise engineers were truly embedded: A number were DRIs (Directly Responsible Individuals) responsible for maintaining particular systems. With this relationship in place, Wise engineers are still the longest-tenured engineers at TpT today - some over a decade by now.
You are more likely than not to need outsourced development at some point in your edtech journey. Whichever combination of in-house and outsourced development teams you choose to build, establishing your needs, staying flexible, hiring carefully, embracing inclusive, iterative development practices, and building a strong team culture will be key to your company’s ability to build good product in the long term.
What’s your experience with outsourced development? We’d love to hear from you.
Edtech Insiders Live Events
Our next Bay Area Edtech Happy Hour is coming up on September 8th in San Francisco! This is a monthly recurring event where we come together as a community to connect, collaborate, and enjoy a drink together! Anyone is welcome to attend.
We’d love to see you there, please RSVP if you plan to attend!
Top Edtech Headlines
1. White House and Tech Companies Respond to K12 Cybersecurity Needs
In response to increased cybersecurity attacks directed towards K12 schools, the White House is launching a new effort to strengthen cybersecurity in schools. This includes $20M from AWS to support cybersecurity training and incident response, cybersecurity support from Cloudflare, courses, training, tools and resources from PowerSchool, an updated cybersecurity guidebook from Google, and cybersecurity courses from D2L.
With the growing presence of AI and technology in schools, we can both appreciate the proactivity of prioritizing cybersecurity in education now, while also recognizing that there will have to be way more to come to fully address this need moving forward.
2. Edtech Industry Funding Update
According to Crunchbase, not a single education technology company has raised a venture round of $100 million or more this year. There is even a question of whether Edtech funding has “hit the bottom of the market.”
While we’ve seen this trend growing over the past year, with marked layoffs and lack of funding, the industry isn’t dead: the most funding traction in Edtech can be seen in AI-enabled edtech tools and platforms that help fill the staffing shortages plaguing K-12 education in particular.
Edtech isn’t going away, but many VCs are keeping their powder dry until they really get a hold on what’s next (and likely jumping less quickly at new tech like Web 3.0 and VR). We’re watching closely to begin to understand and predict what the next era will look like.
3. Chegg, Quizlet, and Flatiron Launch New AI Tools
New AI in education tools are still coming out faster than we can keep up. In the past few weeks, the following big tools have been released:
Quizlet launched a series of new AI features to support their already existing study tools. Quizlet is fast becoming an industry trend-setter on the rapid release of smart and useful AI features; this is likely due to a combination of a fast moving eng and product culture, a massive set of users to test on, and a large amount of proprietary data that can be used to train or tune LLMs.
Chegg partnered with Scale AI to develop a new LLM that will offer more personalized learning and study tools for students.
Flatiron introduced an AI-powered chatbot designed to enhance learning experiences across their online courses
4. AI Literacy + Implementation Initiatives in Education
AI in education, once approached with complete skepticism, is now becoming an accepted part of the era’s technological landscape.
In response, there is an uptick in high-profile initiatives to educate students, teachers, and education leaders on how to effectively and safely leverage AI.
Four big names in education- Code.org, Khan Academy, ISTE, and ETS are working together on a set of resources about AI that can be used for educator training, leveraging videos with AI superstars like Sal Khan, Mira Murati, CTO of OpenAI, and Cristóbal Valenzuela, CEO of RunwayML.;
Stanford has also developed classroom-ready AI resources for high school instructors, and we’re seeing more and more tutorials like The Wharton Interactive’s School’s Practical AI for Teachers and Students coming out, featuring AI education champion Ethan Mollick.
AI in education appears to be here to stay, and educators and student’s alike need support to implement these tools appropriately.
This Newsletter Is Sponsored by Magic Edtech
Magic EdTech has helped the world’s top educational publishers and Edtech companies build learning products and platforms that millions of learners and teachers use every day. Chances are that you're probably using a learning product that they’ve helped design or build!
Companies like Pearson, McGraw Hill, Imagine Learning, American Museum of Natural History have used their help to design or build some of their learning products. Now, Magic wants to bring its pedagogical and engineering expertise to make your key learning products accessible, sticky, and well-adopted.
Recent Edtech Insiders Podcast Episodes
We’ve had some amazing interviews on The Edtech Insiders Podcast in the past couple weeks! Make sure you check out our most recent guests on the podcast including:
Kristina Ishmael, Deputy Director of the US Office of Educational Technology, on AI and national edtech policy.
Dr. Martyn Farrows, CEO of the global leader in voice AI technology for children SoapBox Labs, on unlocking the potential of voice AI in education.
Ben Whately, Co-Founder of memory and language learning app Memrise, on how our brains acquires language and how AI can help
Funding, Mergers, and Acquisitions
For a full deep dive on the latest funding, mergers, acquisitions, and more check out Matt Tower’s ETCH Newsletter.
Funding
Acquisitions
Partnerships
Layoffs
Edtech Insiders Staff Picks
How AI is shaping and personalizing education and learning experiences in Singapore
Shockwaves & Innovations: How Nations Worldwide Are Dealing with AI in Education
How Publicly Traded Higher Education Companies Are Performing
The newest form of school discipline: Kicking kids out of class and into virtual learning
Thanks for reading Edtech Insiders! Subscribe for free to receive a newsletter in your inbox twice a month and support our work!
 | A guest post by
|
 | A guest post by
|