So at the highest level for us in terms of priority is to continue and develop the portfolio of products that is going to serve these markets and put us in the leadership or keep us in the leadership position that we and we see this situation and the modeling that we give also as a transient that is not sure how long it will take and when it will happen, but in a market that is attractive, that has long-term value and that we are very well positioned to lead. And that includes the answer to the question that you have. Where we see the potential to strengthen, accelerate, or improve our differentiation along those surge markets, we will use also our balance sheet to do so, definitely.
Mark Strouse: Thank you.
Operator: Our next question comes from Julien Dumoulin-Smith of Bank of America. Your line is open, sir.
Julien Dumoulin-Smith: Thank you, operator. I appreciate it. Good afternoon, team. I really appreciate the opportunity. Look, just to kind of circle back here to square one. Can we recap a little bit about how we got here, right? I know we’re talking very specifically about the trajectory of the recovery. But can we step back a little bit and understand. You talked a moment ago about June 23 being a high watermark. But obviously there was some insider transactions during the summer. And then obviously things rapidly deteriorated. How do we get confidence on the outlook here? Can you walk us through a little bit more of the sort of month-by-month playbook from that June 23 watermark to where we are today, if you will, through the course of the year?
Obviously, how it gives you the confidence in 4Q, but just how this happened so swiftly? It’s such a pace that we didn’t see this, or presumably you guys didn’t identify it earlier and it’s coming. Again, I just really want to step back and kind of highlight that and how swiftly the cycle moves. And what are the parameters that perhaps — the lessons learned that you guys would identify today after what sort of transpired here, if you don’t mind? And I appreciate your thoughtfulness around this.
Ronen Faier: Yes, Julien, thank you for the question. So I’ll start maybe by going back to Q1, because we need to understand that again, this pattern that we see is coming after 2022 that was a COVID year. And when we’re trying to analyze the business and see where is it going, we’re looking at I would say three major sets of data. The first one is, of course, how much we’re shipping and where we’re shipping it? Second is, what is the point of sale data that we see coming out from our distributors? And as mentioned by Zvi, it’s something that we measure on a monthly basis. And then we’re looking at the inventory levels that are basically a result of these two things. Up until the very beginning of Q1, we saw that our shipping is almost one to one correlated and merged with the point of sale data that we saw from our distributors.
So that means that on one hand, you saw very little inventory days into channel if you recall from our calls at that time, we said that in some cases, we saw close to zero inventory of three phase products, and we sometimes saw one or one and a half months of single phase products. And this is something that we have looked at consistently. At the beginning of Q1, we started to see a relief when it comes to our ability to manufacture single phase products. And we saw gradual relief that actually materialized in full only in Q3 this year in our ability to provide three phase products, especially the commercial products. So what we saw is that during Q1 and into Q2 and Q3, we came back to the normality of a growing market where our shipments into the channel are lower than the — sorry, are going up higher than the actual point of sale.
But when we took the two ratios, our sales into the channels and the point of sale data, we saw that the overall days inventory on hand are ranging between 60 to 90 days, which is the normal level that we would usually see. And by the way, it depends based on the distributor. Some of them would like to be at the higher level because they’re less efficient or more distributed, and some will be more tight. But we started to see a very nice correlation between the two. We saw very steep increase in the point of sale data, sell-through data, and therefore we felt comfortable with the numbers that we’re shipping, and we continue to monitor all the time the inventory on hand days. Now when we’re getting the reports on a monthly basis, usually it happens around the 20th of the following month.
So I would say that when we started July, we at the 20th of July approximately got the June numbers from Europe that were significantly higher than everything that we’ve seen before. These were record high numbers. And as such, our customers continue to make orders or to accept orders, because the backlog was already there. If you remember, we had a very large backlog. Comes the end of August, we are receiving the July data, which is lower than June. But this is again typical because we see a lot of vacations happening in Europe around this time. And we see a decline, but by the way not a huge decline. But we do see a decline. Usually, August numbers are either flat or I would say relatively similar to the ones that we see in July. So comes the end after the 20th of September that we’re getting the August numbers from Europe and the United States.
And we see that August is down compared to July, again, not dramatically down, but it is down. But to their time, this is very end of September, our distributors start to see, and by the way we do not but they start to see that September is not materializing as they expected. And usually the pattern that we used to see over the last I would say several years is that usually you see a slower July and August compared to June. But then you see a jump in September. Actually, we were told by them that the expectation is to see a very strong September. And they acted upon this. But when they started to see that September is not materializing, this is the last I would say two weeks of the quarter, we saw an increasing declines of orders and a much more pressure to delay orders.
And when we received actually close to I would say again 15 to 20 of October the September data, we actually found that September for the first time for many years was lower than the actual August numbers. And this is the area that was a little bit surprising. Now the result is and this is what is leading to our Q4 numbers and guidance is that again, when you take the inventory levels that you see that we’re continuing to grow up again because everyone until sometimes the middle of August expected to see relatively a similar behavior. Now you see that the inventory on hand days are very, very high, much higher than the 90 days that usually are in the channels. And this is when we understood together by the way with our customers that December can — the fourth quarter cannot look even like the one that was Q3 because of this decrease.
The data that we’re not waiting for, and again we do not have yet is how October looks like. We believe that it’s going to be better than September. I do not have any data to see how better it is. But this is the basis of our cautiousness. So as you can understand, this is a pattern where the deviation from the pattern that we used to see was something that was actually I would say understood in the two to three last weeks of September, and not before. And that led to basically the fact that a lot of our cancellations and orders came actually at that point of time were push outs. So I hope I answered your questions.
Zvi Lando: Maybe, Julien, on the topic of course — the dynamic is as Ronen described from it, but we need to learn from it. One of the elements that is meaningful that also we’re talking with our distributors and how to handle is visibility into the inventory levels at the installers, because what’s happened as part of this dynamic is we’re coming off of an unprecedented event in terms of the surge in demand, the lack of availability, everybody was building on inventory, including installers that typically don’t take on inventory, and especially inventory of products they prefer to install. Then when they slowed down, they slowed down their orders but they had inventory and they consumed it and that was not visible. So in terms of data points to add to the data that we typically track of installation rates fell through inventory levels at our distributors is inventory levels at our large installers, that will help have better clarity when fluctuations occur, especially such extreme fluctuations.
