Erik Suppinger: Yes. Thanks for taking the question, and congrats on a good quarter. So, one, you said that the churn is about what you had expected. Can you give us any sense for whether you were expecting a change or how much you think it does change, from what you’ve seen so far? And then, two, can you remind us what your performance was relative to AWS before you made the enhancement? Were you at parity and now you’re 30% faster? Or where — what was the relative change, and do you think that you’re going to continue to improve your relative performance to the S3 performance?
Gleb Budman: So, Erik, this is Gleb. On the performance side of it, we really focused on measuring our performance versus Amazon once we had ChartSlash [Ph] put in place. So, we think we were probably a little bit on the slower side before, and now we’re quite a bit faster than they are at this point. In terms of can we continue to innovate further down the performance side, the answer is yes. We have components in the roadmap to do more of that. We’ll see whether — to what extent those continue to, make the scale of those improvements, but we do have items that we see as possibilities for us to continue to drive further performance improvements. And then on the other question that you had, I believe — you faded out for a second, but I believe it was a question of how was — what were our expectations around churn from the price increase?
Erik Suppinger: Yes, that’s correct.
Gleb Budman: Yes, and we were — we had history on the computer backup price increase, although this one had our extended version history being bundled into the product, so you’re actually getting a product innovation on top of, and then, associated increase to that. So, the — we were modeling in just a modest — more modest there, and we are — and our B2 was — we modeled in a churn rate, and so far, we are better than that model churn rate, but it’s very early, and that’s why we didn’t want to say more about it yet.
Erik Suppinger: And then coming back to the performance question, is this the type of thing where S3’s performance is steadily improving over time, or is it a fairly stagnant kind of performance metric that you’re — that — on the S3 side?
Gleb Budman: So, I would say I’m not sure that we have good data on that. During our testing, it was very stable, so it’s not the kind of thing where it moved around wildly during our testing, nor did we see any performance improvements on their front during that period, but that period was fairly short in terms of how S3 performance compared this year to last year to two years ago. I’m not sure that we have that data. Maybe there’s some public information available on that. But in general, I think they — what they’ve been doing is S3 has kind of been the default offering, and then they’ve had slower versions, things like Glacier, which are slower, but not necessarily things that are faster. So this is currently faster than their top tier of service.
Erik Suppinger: All right, very good. Thank you.
Operator: Our next question comes from Eric Martinuzzi with Lake Street Capital Markets. Please go ahead.
Eric Martinuzzi: Yes, and the renewals you talked about, at least based on the one month of evidence that you’re seeing customer, the churn in line with expectations, and then kind of the macro perspective, we’ve got data growth, but there hasn’t been any deceleration there. On the computer backup side, at least for the one month of observation, have you seen any pushback on people asking for terms on their one or two year, I know it’s a one or two year upfront payment, any pushback there?
Gleb Budman: It’s funny as we’ve actually seen some customers wanting to switch from monthly to one and two year to lock in the price point ahead of the pricing.
Eric Martinuzzi: Yes, okay. So the dollar amount’s not an issue for people. Okay. And then shifting over to the 2024, I know you’re going to give us a full year outlook in February, but one of the things that we’ve seen here in 2023 is on the gross margin side is some data center investments. How should we be thinking about that for 2024?
Gleb Budman: Yes, we have always said that gross margin was going to be in the mid 70% on the non-GAAP. And now we think of it as in the upper 70% range.
Eric Martinuzzi: Got it. Thanks for taking my question.
Operator: Our next question comes from Zach Cummins with B Riley Securities. Please go ahead.
Zach Cummins: Great. Hi, Gleb. Hi, Frank. Thanks for taking my questions. Just starting off, Frank, can you talk a little bit more about the cash usage? I know it sounds like it’s supposed to fall by about half in Q4. I think you gave a target near the end of August of at least 20 million exiting next year. Are you still feeling pretty good about that target or how should we think about cash usage?
Frank Patchel: That we provided last time was that we would exit 2024 with $20 million and we still feel very good about that. And that the cash break-even would be in the first half of 2025. So that remains the same.
Zach Cummins: Got it. Helpful. And one question for Gleb, just around the opportunity with the AI training models, interesting to see kind of your closer partnership with the Coreweave. Can you just talk about the potential opportunity you’re seeing there and any potential tailwind you could see as more people look for ways to train AI models?
Gleb Budman: Yes, absolutely. I mean, I think we’ve all seen that AI uses data at large scale and that the growth in data and data needs and data usage, data evaluation is growing exponentially. We’ve shared customers and how their use cases are on this call and prior call and others that where our customers are using our storage to — keeping the data that’s being used. So Backblaze B2 Cloud Storage has a play in different parts of the workflow for AI. And I think it was interesting that in the example that we brought, that we shared of the customer, they were a Coreweave customer that started looking for how do I find a place to keep my data somewhere where it’s affordable to keep it, but also easily accessible to Coreweave.
And so Coreweave actually introduced them to us. We’ve had customers that have come to us off of some of the traditional cloud providers and looked for it, used us first for the storage and then we’ve introduced them to Coreweave for the compute side of it. So I think the overall opportunity is data growth in lots of different areas as a result of the AI use cases, the usage of that data increasing and the fact that you want your data to be accessible to the different places where you want to run models and do analysis and that therefore it needs to be part of this open cloud ecosystem. That if your data is locked inside of a traditional cloud like AWS, then it becomes expensive to leverage it with other providers like Coreweave and others where you want the data to be accessible to them.
So I believe that that position as an open cloud provider for storage is an increasingly good place to be when customers want their data accessible to all these different model providers.
Zach Cummins: Got it. Well, thanks for taking my questions and best of luck with the rest of the year.
Gleb Budman: Thank you.
Operator: Our next question comes from Jason Ader with William Blair. Please go ahead.
