Interested in Meta Releases Llama 2 Long? Check out the latest video and resources from Tyler Bryden on Meta Releases Llama 2 Long!
This is is part of my live-learning series! I will be updating this post as I continue through my journey. I apologize for any grammatical errors or incoherent thoughts. This is a practice to help me share things that are valuable without falling apart from the pressure of perfection.
It’s important to self reflect. While we continue to grow, I think all of us on the team don’t feel like we are living to our full potential.
This is definitely a super critical look at everything.
This is my fault. I’ve muddled the message. I’ve always wanted to build a consumer product that everyday people love using. I’m getting closer to that mission. But, it’s been harder to get there then I thought. So, I’ve had to try to make some sacrifices and build the system so it works really well for organizations too. That leads to confusion and feeling the product isn’t tailored to anyone.
I am still trying to figure this out. I still can’t even describe exactly what we’re building most of the time which is not great. It’s almost like a new digital tool for modern day professionals also trying to improve their personal life.
When I look at it overall, people are the ones who use products within an organization. If I can’t make it good for one individual then no one in a company is going to use it. So, I do have this parallel mission. But, it’s not easy. We’re suffering because of it.
No fear of missing out. No one believes we’ve hit something yet that they just can’t miss out on. That’s what we are seeing with Roam Research. People scared if they aren’t participating and using that software they are missing out on key insights into themselves. People are paying for it without even using Roam because they’ve seen others discussing the benefits. Wild. Impressive.
We need to create that feeling. And it needs to not be manufactured but true.
We do not have a talented UI/UX designer on the team. This means we are doing a lot of eyeballing. Surprisingly, the application is decently intuitive, but it is not inspiring.
The technology foundation is laid we just need some support and we really think this could make a huge difference.
However, it also highlights a potentially fatal flaw: we haven’t done the research from the start, leading to tech debt. Leading to features that no one is using. Leading to silly bugs and bottlenecks for users.
Some people speak highly of us and share things online and on LinkedIn, but I believe I’ve muddled the message so much people aren’t always communicating it the right way. I would say again, the brand and the messaging is not strong enough and this a result.
Until people start using the product and organically sharing their experience in a positive way, we will struggle to grow.
We haven’t dedicated a ton of time to engaging on social media, which in this age we all know is so important.
I need to stop making excuses and just start to interact more. I do a lot personally but I don’t do much through the company lens. Sometimes, that is because I feel like it doesn’t fully align with my identity, but that’s okay, and I’ve got to get over that.
We do have some resources here:
But, it is obviously not enough. We need to cover:
How to use the platform? What is the value?
This came from some research I did about organizations not paying for a tool. They need a solution to their problem. We are not there. We’ve tried to solve too many problems leading to a bunch of partway solutions. Until we can hammer home and solve one real, painful, and urgent problem, we will struggle to grow.
As one example, our transcription workflow is not good enough. Our transcription is not accurate as it needs to be. Sometimes our punctuation fails. We don’t have word-by-word time-stamping. You can’t edit speakers within the application. Our transcript editor isn’t strong enough. There is no human transcriber option if you don’t want to clean it up to 100% yourself. These all screw up our selling and growth.
We have people signing up on the platform all the time but there is no feedback loop or strong communication.
When someone signs up, there is currently one onboarding email set up. We often don’t really know who they are, how they found us (besides through Analytics data) and we never actually hear from them. I thought getting in touch with people who sign up for your app would be a lot easier but it’s surprisingly hard 😂
There are other companies killing it without these, so it’s just another excuse.
But, this would be a great way to increase our daily usage and also drive a new customer adoption channel. Right now, both of these are in development, but on top of everything else, it’s been a difficult journey and none of us are happy with the current iteration.
We hope to release at some point but there is work to get there. And with those releases, a whole other new challenge of doing a good job there emerges.
If for example, you have been note-taking in Evernote for years, it is difficult to make a switch over to a new platform. There is so much value in the notes you’ve made in the system, and over time, it becomes a “Second Brain” that is hard to leave.
This is why Notion put so much work into and advertises so heavily their Evernote Importer. There is a whole page dedicated to it.
Until we make importers or syncs with the most popular platforms, it will be difficult for people to start using Speak.
Just as an additional note, we don’t have the desire to replace all the functionalities of Evernote, for example, importing images, which would be a dealbreaker for people and stop them from using us.
If we look at competitors like Notion, they have started to offer free or $4.00 a month plans. Roam Research is an outlier at $15 USD per month. Most people wouldn’t allocate that much budget to a productivity tool. It’s a specific group who does. It’s good, because it narrows the target market. But, now with Roam cemented in a lot of people’s mind (people paying $500 USD for their believer plan on a 5 year timeline), there is a decreasing amount of people who would double up and also pay for our platform. At least, that is what I am assuming.
For consumers, we do have a free plan. That plan doesn’t include any audio and video though besides the 30 minutes we give you when you sign up. Compared to Otter.ai for audio transcription, and Vidyard for video, we are lacking and unimpressive.
We are not venture funded so we haven’t been able to just give things away for free. Our servers, our processing, it all costs us.
This comes from the top down. I am interested in too many things. A lot of the problems I am trying to solve are detached from the every day problems others need to solve. Example: how can I capture a psychedelic experience with an app considering people often start in one place and walk all over etc etc. Who cares? And who of those people cares is actually going to pay $19.99 USD a month or more to solve that. In my opinion, not many. I have to be honest with myself. It’s great to scratch my own intellectual itch but that doesn’t build a business necessarily.
A powerhouse. Already had dictation. Not too long before they integrate their own speech-to-text better into their suite of productivity tools.
They have so much money, so much capacity, and a system that billions of people are using already. How and why do we want to compete there? It’s an easy choice for users on Google over us.
It’s the same thing with Microsoft. They just added transcription to the web version of their Word Doc platform. It won’t be long before it’s on the desktop version too and people will be taking full advantage of their speech-to-text at little to no cost.
The OG. Everyone doing knowledge management has most likely used Evernote at some point. Amazing full-suite of features. I believe up to 250 million users. However, they only have a 1% to 3% conversion rate to paid users. And even then, those plans are small.
That’s with them constantly giving out offers too. Check out the offer I got when I just visited their site to see their plans:
What does this tell me? People love taking notes, but maybe not enough to pay for it. It doesn’t seem valuable enough in our life.
That’s the idea I’ve been thinking about around “noteworth”: how if you take notes properly, just like investment, over time the value compounds.
A force and leader in the space that has disrupted the longtime incumbent Evernote. Beautiful interface, the ability to extend capabilities, simplicity and a thriving community has made Notion a mainstay for productivity people.
Roam has come into the scene like crazy. People proudly exclaim they are part of the “#RoamCult”. It’s fascinating how veracious their users are in exploring the platform and promoting it to others.
Take a look through Twitter and somewhere you will find someone talking about Roam, replying to a tweet when someone asks about a note-taking platform. It is incredible.
This platform will continue to grow with the momentum it has. Already, they are over $1 million ARR. They’ve taken on some investment and are just deep in a niche that will continue to explode. That niche can be considered “Tools for Thought” or “Building a Second Brain” and a lot of the excitement comes from the graph database and bidirectional linking which I will explore more.
Waterloo company. Focused only on video so there is some differentiation here. But, they are still a force. A lot of the times people see our platform and ask about VidYard. “Have you heard of VidYard?” “VidYard would love this feature?”.
Wistia is along the lines of VidYard. They have a great brand and seem like awesome people.
A full suite of image, audio and video tools. It’s inspiring how well they’ve done their work. They’ve got a great, growing team with amazing content.
Second largest search engine. Built-in audiences. Automatic free captioning and translation. Embeddable players. Now added chapters for easier navigation of content. Such a force. Almost impossible to compete on many areas.
Great, fun platform for recording videos for free to share. Blowing up like crazy with another $30 million USD injected into their company. Adding transcription and captioning. So much potential.
One of the companies that when they launched, it made me want to quit. At the same time it was exciting: they were the first company I felt that shared my vision and the power of capturing conversations.
Their application is ridiculously good and from my test far exceeds in accuracy. They’ve built their own speech recognition system from scratch which enables them to be more profitable. Their plans are cheap and give you so much power and time:
Frankly, we don’t have the money, technical knowledge, or capacity to compete with them.
Sometimes even more worrisome, I feel like they have the power to replicate what we are doing. For example, they don’t have video right now, you only do audio which is differentiator for us. But, at any moment, they could just embed the video file people are uploading for playback and there’s a huge hit.
They only extract keywords right now, but they are not that far away from doing named-entity recognition like us and extracting people, names, locations and more.
Built by ex-Googlers, they are a force to be reckoned with. They have an amazing relationship with Zoom which has skyrocketed their business. They will continue to grow and dominate. We’ve bumped up against them many times, and although we’ve won several of those for some of the features we have, it’s a challenge that will continue to arise.
Killer company. Focused mostly on sales. Now branching out to other areas. Already worth over $2 billion USD. They are just beginning. Their insights are ridiculous. I see them being a massive disruptor and will start to bleed into other areas and markets that we are sitting in.
Similar to Gong and focused on sales. Also experiencing high-growth. Been around for a while. Strong brand. Releasing some great research and information products. I think they will have a tough time with Gong competing directly with them but it seems they’ve still got some valuable customers.
This company doesn’t get mentioned as much, but they are doing some awesome work and I’ve heard it pop up a few times. More focused on transcription, but they’ve got a good brand, continue to release updates, and have a good core offering.
Trint is a great brand. I haven’t heard as much from them recently, but they’ve got a core customer base in journalism, reporting, research, and media companies. They’ve injected capital into their company too. They are Toronto and UK based with an awesome CEO/founder.
They are inspiring. I’ve learned from them. I will continue to. There is no doubt that as people find our option and explore others, Trint will come up and we will need to have answers why our offering is better.
Speak Ai is decent. But, the logo on our site is just Speak. Some people call us Speak Ai others call us Speak. A lot of people spell it SpeakAI. Some people put a small “i” others do a capital “I”. It’s honestly a mess and I hate it.
Until we find a name that is spelled and pronounced one way and also encompasses what we do, we could be lost. Maybe I am making that up in my head but it’s definitely a thought.
So many companies are offering transcription/speech-to-text. This space is too competitive. If we are to just focus on that we will not be successful. On top of companies like Otter.ai and Trint, we also have cloud services on Amazon, Google, Microsoft, and more. These companies will continue to improve their systems and anyone who can develop can tap into those systems easily with a few API calls and build something functional.
We don’t have that much content. And because what we are doing is somewhat abstract, this just becomes more and more clear this is a huge issue.
Because we lack content, we lack domain authority. We lack a social media following. We don’t drive enough awareness and traffic. And because of that, we can’t convert enough people into free users, which means we don’t have enough people to convert to paid users.
If we use Evernote as a benchmark at a 2% conversion rate, we would need 1,000,000 free users to get to 20,000 paid users. If that is the first level of the plan at $19.99 USD, we could get to $399,800 per month in revenue. That doesn’t seem as scary as I thought but we are still a long way off from getting there in my eyes unless we can make some dramatic changes quickly, or some virality or feedback loop takes off.
Besides a niche Invest in Ontario and MSN article, we haven’t pushed our brand to the world. A lot of that comes from feeling unready.
We haven’t been Product Hunted, we haven’t been featured in anything. Do we have a story worth telling yet? I’m not so sure. Also, I’m still not satisfied with Speak/Speak Ai. Until we find something that fully encompasses what we are doing I’m not sure I’m ready to make that push. And until we do, it severely hampers our growth potential.
This is just a reality. We can only do so much. We’re moving lightning quick, but that doesn’t compare to the armies that other companies have.
All 4 of us bring something wonderful to the table, but that doesn’t mean we are capable of moving mountains (yet). Unless we figure out how to get people to make big commitments to paying for our software, or we take on investment, it’s going to be hard to scale up the team to compete.
This is something I personally struggle with. I am not that much of a Capitalist. I want to sell but I am not trying to force our solution into a place it doesn’t fit. I am honest about the shortcomings of our application and don’t want to under deliver. A lot of these are mental blocks I need to get over. I am watching people with inferior products to ours generate millions in revenue each year.
I need to turn our company into a sales machine even if I don’t like it.
We’ve bootstrapped this to date with revenue from consulting and some grants and hiring subsidies, on top of some relief measures from the government because of COVID-19.
There is a bit of a bottleneck here, because we believe we know what we need to do, but we don’t have the funds to make that happen. Now, I am writing this and there is a voice in my mind yelling “bullshit Tyler”. That is true. It’s about making the impossible possible, and if I can’t do that, then I’m not the person to make this successful.
But, still, I can be realistic about the limitations. A lot of technology companies feel this pinch and it’s difficult when you’re not just making a simple product or technology but a large-scale, complex offering with machine learning and Ai.
There is no doubt about that. We love what we are doing. Anyone who meets us can see the passion and energy. We’re not always putting it in the right places but we’re trying. That passion has driven us through a lot already and will continue to.
Although it may seem like I’m complaining here, it’s been wild to see how far we’ve come from the beginning. We actually do have a valuable platform. Whether it’s been personal goals, company goals defined by ourselves or programs, we’ve kept up a incredible pace and hit a lot of goals that seemed difficult or impossible at the time. We hope to see that continue.
Voicetech. TedX. Psychedelics. Mental health. Neurotech. London Ontario. Toronto.
We haven’t grown the quickest sometimes because our whole team is engaged and cares about communities.
We aren’t looking for these to pay off for our company, but it seems the more we invest in others the more good comes back to us. These relationships and reputations are based on trust and doing the right things, and we will continue this.
Focus on transparency within team, to our users, and with the world at large.
That’s why we put things on IndieHackers. Here’s the link. The page shows off all the milestones we’ve hit, our current MRR (monthly recurring revenue) and more.
Ontario Centres of Excellence, IRAP and SR&ED have been so kind to us. They have supported us with advising but even more then that, financially to help us build and grow.
These relationships and trust we’re hard to build but make us strong and resilient.
The DMZ has been an incredible force. Even with the pandemic, they have guided us in a positive direction, introduced to investors, customers, legal teams, and more. We can’t thank them enough, and it again proves there is strength in community.
I can’t say we’re entirely helpless here. Otherwise, we wouldn’t be here. We wouldn’t be growing. We wouldn’t have had almost 250 people sign up. There is something that is resonating. It’s just not perfect, or even that great. However, I will try to list out some differentiators:
These are three powerful mediums I believe in so much. We use them a lot as people. We have built a platform that can capture, analyze, and distribute those all together. It dilutes our focus, but at the same time, for example, Vidyard only does video. That is great, but you can’t write text notes. You can’t do audio in the same way. In one way it makes us weaker because it’s hard to do all really well, on the other hand, if we execute, this is an amazing differentiator.
One more example: Otter.ai is great for transcription, but if you want to edit it and turn it into an article, you need to leave the platform. With our platform, you can convert the transcript into our text section, immediately edit it, and then post that as an article etc. We can break the constraints of transcription which several platforms don’t allow you to do. Now, again, that may make us weaker: maybe Otter.ai thinks Microsoft Word and Google Docs are already good enough and there is no competing, so instead of trying to, we’ll just do nice export formats and enable users to work with those systems. They can’t do everything perfect.
Still, I think there is something here.
We are focused on extracting insights from this data, and for example, not inherently transcription itself. Speech-to-text is just a vehicle for getting to our analysis and insights extraction.
I think this is one of the innovations we are bringing. Instead of just leaving every note as an individual note, we are combining all the data together over time periods to aggregate and produce interesting insights. There is still a lot more to do here, and again, we run the risk of competitors doing this, but I think we are approaching this in a unique, valuable way that could be beneficial to users. Who are those users and what insights are worth surfacing? It seems we still have some work left to do to figure that out.
Even though the team is small, we are deeply technical. Both conceptually and in development.
We love technology. We love solving problems.
This is so cool. Technical limitations to making this full functional at the moment but it’s just a snippet of what we are thinking of and what we can build with the foundation we’ve laid.
Real-time information processing will continue to grow. We want to have better human-computer interaction. We want to be part of that and have done a lot of thinking, research, and development to make that happen. A lot more will come.
I’ve made it a goal to reach $1 million in revenue 2021.
You can find out how I am trying to achieve that at the link here.
Thanks so much for checking this out. I’ll be back to improve this content for you.
Interested in Meta Releases Llama 2 Long? Check out the latest video and resources from Tyler Bryden on Meta Releases Llama 2 Long!
Interested in Canada Voluntary AI Code Of Conduct? Check out the latest video and resources from Tyler Bryden on Canada Voluntary AI Code Of Conduct!