Storj Town Hall - January 29, 2025

Storj Town Hall - January 29, 2025

storj-townhall-forum-012925

Learn about our latest achievements, highlights, and what to expect from Storj in the coming months.

We hold Town Hall sessions regularly throughout the year to keep our community up-to-date on all the latest from Storj, our token practices , and to give the Storj Community the opportunity to ask questions and learn more.

Follow our YouTube , Twitter and LinkedIn accounts for the latest Storj news, updates and events.

If you have questions about the January 29, 2025 Town Hall please ask them in the replies below.

5 Likes

What is the strategy to 1 EB of customer data storage in 2030 ?

4 Likes

@jammerdan We are very optimistic about our future but that is too far out to forecast at this time

4 Likes

With all tranches unlocked, is it likely Storj will need to source some tokens from the open market this year? Or is there probably still enough in the treasury to make it through 2025?

5 Likes

Have you procured new microphones for the Town Hall today? :slight_smile:

8 Likes

I hope you can provide the reasons and Numbers why you are so Optimistic for the Future in the town hall.

Thx :slight_smile:

1 Like

Your optimism about the future is great, maybe I can join back in again after listening to the Town Hall today. But anyway I really hope to receive an answer: If forecasting 5 years ahead is too challenging for Storj, I’d be happy to hear the timeframe that the company is happy with to share forecasting, goals and plans. What is the Storj goal for the amount of customer data storage by the end of 2025 for example?

With my mentioning of 1 EB in 2030, my intention was to make clear that a goal like this inherently requires a strategy and planning. This involves breaking down the objective into actionable steps to achieve it. To point out what such an overall objective would mean broken down into shorter term e.g. yearly objectives: Achieving 1 EB in 5 years means an average of 200 Petabytes of additional customer data per year. This is massive when we look at the current quantity of stored data as of today. To reach such a goal or just getting even close to it, it appears to me that significant changes would be required.

It would likely require a multi-faceted strategy, including substantial investments in sales and marketing, maybe even additional acquisitions or a fundamentally different approach to markets. That would be no longer a matter of just incremental growth, but rather a transformative effort that would require a strategy, careful planning, execution and resource allocation.

Long term goals are not unusual by the way: In the context of enterprise business, strategy planning typically involves setting long-term goals, often with a 3-5 year horizon, and then derive shorter-term objectives which lead to actual actions. Such planning will allow to allocate resources, invest in infrastructure, and make informed decisions about the future direction. Given the complexity and scale of data storage, it’s reasonable to expect that Storj have already begun planning for the next 3 to 5 to 10 years.

It’s worth noting that Storj competitors traditional data centers or other distributed storage solutions likewise are not sleeping. Companies such as Wasabi, have already planned and built exabyte-scale storage capacities. This level of investment and planning requires a significant amount of forecasting and strategic planning, often 5-10 years ahead. If companies like Wasabi can plan and execute on such a large scale, it’s reasonable to expect that other players in the industry, like yours, have also begun to think about their long-term strategy.

To put 200 PB per year into perspective, the total amount of data created, captured, copied, and consumed globally is expected to reach 175 zettabytes (ZB) by 2025, according to IDC’s Global DataSphere Forecast. This represents a significant increase from the 64.2 ZB of data created in 2020. Moreover, the amount of data stored in the cloud is expected to reach 100 ZB by 2025, growing at a compound annual growth rate (CAGR) of 23.6% from 2020 to 2025. We see already 394 Zettabytes of data/information created, captured, copied, and consumed worldwide for 2028 (Data growth worldwide 2010-2028 | Statista) and 500 Zettabytes for 2030 (IDC), underscoring the unstoppable trajectory of data growth, which could mean with the same ratio applied that we are talking about approximately 300 ZB of data stored in the cloud by then.
The overall question is, what share of this does Storj want, what is reasonable and finally what is required to reach it?

Exabyte storage capacity had already been a topic for Storj as far back as 2022 when on https://forum.storj.io/t/lets-talk-about-the-elephant-in-the-room-the-storj-economic-model-node-operator-payout-model
@John had stated

Or as @Knowledge has said:

Back then it sounded like Storj has a vision, a goal and a plan. The current results are modest at best in my opinion. I believe it’s essential to have a clear strategy in place and clear goals to bring up the most suitable actions to address the ever growing demand for cloud data storage. To put it with @BrightSilence

8 Likes

I’d prefer to think about this as 6.2% growth month-to-month in the amount of paying customer data stored. Is Storj on this trajectory now? According to the last Town Hall, month-to-month was at 11%.

4 Likes

I think you can summarize all of this by just stating, “Storj needs more customers to store data. What are you doing to get more customers to store data?”

We don’t have hundreds of millions of dollars to spend on massive advertising campaigns. The marketing team advertises and gets articles in trade publications as well and the company attends various shows where it makes strategic sense to do so. There is an operational budget like any company, and the team has to operate within that so that the runway the company operates on is comfortable for Storj to continue for years. Such things are important to customers when making sales.

We have new products that customers are interested in as well. These products put customers at the sales table and allow them to see all of our offerings, and not just the ones they initially inquired about. They may not have been thinking of storage when they reached out to us, but they may start thinking about it once they’ve met with us and seen what we bring.

Indeed, data for SNO’s has been slow and sometimes reversed as they clean out the junk. While that can appear to the SNO that nothing is happening, the reality is that behind the scenes, there is a lot going on every day between Storj and prospective customers, as well as engagement with existing customers. I think if you were to see how sales were done and the process behind it all, you’d have a better understanding of how long it can take to get a customer to move to a new product from their old product. You’d be amazed at how much time businesses can waste stringing along conversations about potentials and then decide to go in another direction. We can have potential customers who want to use the product but may not be ready for a year or more because they are building out something new that won’t be ready until then.

We’ve landed some customers recently that indicate our ability to sign on large customers with complicated procurement processes and specific requirements. The Storj team was able to deliver on those requests. We are taking that momentum into 2025, and I think we’re all expecting great things from the company this year.

10 Likes

I can understand sales taking time. I’ve been part of sales proposals where a potential customer looked at our product… and decided to have their own IT/devs make something internal instead.

And THREE YEARS LATER we come back to them… and they’re running some half-baked/half-built/half-deployed/half-working/half-funded and unsupported tool made by their own people. THEN they decide to buy from us :wink:

I know sites like storjstats.info show “stored customer data” down over the last 12 months… so it can be hard to believe claims like 11%-month-over-month growth. But I choose to believe that growth has been hidden inside a mountain of forever-free/unpaid data that will eventually be cleared out… and the real growth will become visible.

10 Likes

How far is SOC2 for public network?

7 Likes

Many of us said Storj needs to address the personal storage market, and Storj always replyed that it would assume another company entirely. If it’s hard to make a new branch of Storj for this market, is there a plan to aquire such a company, that already exists, and brand it with Storj logos and names?
That will fill the hole in Storj offer, and will bring us closer to personal storage and individuals, and get us more data.
Dealing with contracts for big businesses it seems it’s pretty hard and very uncertain.
Dealing with individuals, you don’t need 1000 pages contracts and 20 lawyers, and you don’t have to spend months in convincing them of the Storj benefits. Individuals just click Accept the Terms and Conditions without reading them, and they just start using the service and pay for it.
We already have a forum for helping anyone in need. So the support is already in place.
You just can’t rely on third party apps to bring you more clients, apps that offer Storj as an alternative among other options.
We need a service and app for personal storage with only Storj public network as backend.

4 Likes

Not really. It was more specific. Of course we want more customers specifically for the public network. But it was not only that. It was about where Storj wants to be by the end of 2025, 2026, 2027 to 2030 in terms of what quantity of customer data Storj aims for and what are the strategic plans and operational actions to achieve this.

I don’t think you need hundreds of millions of dollars to spend on massive advertising campaigns. But I believe you need to invest into localization, regional sales and support and into getting the public network compliant for HIPPA, SOC2, GDPR etc.
If there is lack of funding, well then new investors need to be found, maybe even another ICO. I don’t know.
Go to the Impossible Cloud website, click on “Customers” and read and view the testimonials. Those customer are telling you exactly for what reasons they have chosen Impossible Cloud.
We also see Storj not mentioned very often on Social Media, Reddit for example. Everyday people are asking for good storage solutions. We mainly see the very same answers: Backblaze, Wasabi, maybe even the brand new Hetzner Object Storage. I am just asking myself why after 10 years in service Storj hardly ever gets mentioned.
Marketing by recommendation is the best and cheapest kind of marketing you can get.

Customer testimonials and case studies are another thing. Yesterday Vivint was mentioned in the Town Hall. So obviously they have no reservations to be named. And we were told they are happy with Storj. So wouldn’t that be a great opportunity to write blog posts, case studies about it. Even a Youtube video could be done as this customer seems to be a massive one.

Yes and this is great.

Well Storj sales team is not transparent about it, then we can hardly know it. I can remember (however I don’t find the link right know) that I had asked here several times to present more about the work they are doing and what obstacles they may face. It is really up to Storj if they let us know or not.

We have seen that with Vivint. But not all customers are like that.
I remember that customers were very quick to move massive amounts of data onto Cloudflare Object Storage or Hetzner even during test phase. I can remember posts on social media from users who planned to move double digit PB of data onto it without hesitation. So yes, there are potential customers where it needs more time but there are also others with Petabytes of data who are are quick to leave their current object storage provider, better today than tomorrow.

2 Likes

Your curiosity about what the company is doing and how they are going about it is interesting, because you make a lot of assumptions about what Storj isn’t doing, when they may actually be doing many of those things but you just don’t hear about them. And why don’t you hear about them? Because in many cases Storj doesn’t want their competition to know. The competition likely monitors this forum and uses whatever negatives they find and leverages that against Storj. Even taking your words and saying, “Look, this is one of their users complaining about the company’s lack of these things here, it’s obvious they have problems. Where at the same time, that same person is promoting the things we are doing right. Sign here.” Storj transparency doesn’t mean a complete view of everything happening inside the company. It’s very competitive out there, and detailing the sales process in a public forum is not something any business is likely to do. I am sure if Storj is working with partners, funding managers, potential customers, or anything of that sort, if those questions come up after NDA’s and such are signed, they will go over all those processes with them.

8 Likes

In https://youtu.be/ZoNijqK-GN8?si=IbDSggk6ZWuoK5LD&t=339, the chart don’t show specific number anymore (the last 2 or 3 Town Hall did not show number as well if I remember correctly). But because there was number in video in the past, combine with the visual current chart - it can be deduce to the current number – which is the thing only your competitor would do, and it is a disservice to your loyal community member.

Another example: https://forum.storj.io/t/announcing-community-satellite-pilot-program, and the move from cockroachdb to spanner, it will disable most attempt from opensource/self-hosted community. The one can really afford it – is your new competitor.

Lastly, secrecy doesn’t give you media power, the thing that own the world. Maybe Storj think having enough deal from customers it will start to cascading – I’m not sure, I’ve never run a business. I’m just thinking out loud: business executive understand game theory very well, what do they gain from recommending another service to anyone?

Or, in case of Storj, they’ll set up their own internal satellite with consulting help from Storj. I recall Storjlings talking about projects of this type. Notable advantage of this kind of approach is being able to colocate compute with storage, which is not possible with the public network, and might be necessary for transfer cost or latency reasons.

This also brings revenue to Storj. But will not bring much to node operators :person_shrugging:

For the reason above I wonder if Storj wants to keep satellite support for cockroachdb after all? Or, let say, postgres.

And flood Storj with support requests, which have to be handled. There are trade-offs to this approach.

3 Likes

Both are still supported. I didn’t see the roadmap items to drop their support.
If the backend is implemented correctly it should be modular, so potentially it could be replaced to a nextgen distributed database without much of efforts (maybe except the migration and the transition period).
Who knows, maybe someday there would be a fast reliable distributed database with zero trust and we might use it, or improve our network so far, that it will be no brainer to use as a storage for such a database.