Skip to content

ON - Standup: Investments Dashboard & Scenario Prototype - 2026-02-18

Metadata

  • Date: 2026-02-18
  • Company: ON
  • External Participants: Rodrigo Cabrera, Amanda Mitt
  • Palm Participants: Emma Sjöström, Jennifer, Giannis, Federico Morando
  • Type: Customer Call
  • Domain Areas: Investments & Debt, Scenario Modelling, Connectivity, Reporting
  • Recording: None

Summary

Context

Weekly standup between Palm and ON's treasury team. Two major demos: Jennifer showed the new investments dashboard (nearly ready to replace ON's manual Google Sheet), and Emma demoed the scenario/what-if prototype for initial feedback.

Key Discussion Points

  • Investments dashboard nearly complete — Jennifer demoed metrics: active investments count, weighted average rate, counterparty spread, upcoming maturities, interest earned. Self-serve reporting (custom canvas) now includes investment data. Rodrigo can soon decommission his manual spreadsheet.
  • Scenario prototype feedback (8-9/10 excitement) — Emma showed percentage-based what-if assumptions on categories. Amanda and Rodrigo found it intuitive and relevant. Part of ON's 2026 goals to do more scenario and stress testing.
  • New use cases surfaced: working capital scenarios (AR, AP, inventory), FX currency scenarios, new market entry (Sweden store example — tricky because no historical data exists for new entities).
  • Lego-piece mental model resonated — idea of composing small assumptions into larger scenario "events" (e.g., "Swedish store opening") was well received by Amanda and Rodrigo.
  • Fixing scenarios vs rolling forecast tension — Jennifer raised that once you set a best/worst/baseline scenario, the rolling forecast re-baselines to actuals, wiping out the comparison. Need ability to "fix" a scenario and overlay actuals on top.
  • MMCs need adding to investments — Amanda flagged that their money market-like products (UBS) aren't in the investment total yet. Monthly frequency, interest reinvested as principal.
  • BigQuery/Dynamics API integration — Amanda shared that ON has a new IT product manager for Dynamics who is excited about the integration. Ready to kick off, may get resources in Q2. Wants specs/documentation from Palm.
  • Anaplan budget data — discussion about feeding Anaplan capex budgets into Palm for better scenario coverage.

Pain Points

  • Weighted average investment rate calculation shows zeroes on days with no deposits — distorts the average
  • MMC/fund data not integrated into investment totals — creates incomplete counterparty exposure view
  • New entity/market scenarios impossible to forecast when there's no historical data for that entity
  • Rolling forecast re-baselines to actuals, making it impossible to compare against a fixed scenario over time

Feature Requests & Needs

  • Include MMCs (money market fund-like products) in total investments view
  • Ability to "fix" a scenario and overlay actuals on top for tracking
  • Confidence bands on forecast (80% certainty range)
  • Manual input for scenarios where no historical data exists (new markets, acquisitions)
  • Intercompany and non-business categories in scenario assumptions
  • Investment data included in scenario scope

Jobs & Desired Outcomes

Job: Stress-test forecasts with what-if assumptions without leaving Palm

Desired Outcomes: - Minimize the time required to model impact of business events on cash position - Increase the ability to combine multiple assumptions into coherent event scenarios (e.g., new store opening) - Reduce dependency on Excel for ad hoc scenario analysis - Minimize the gap between scenario creation and actionable insight

Job: Track total investment exposure including all instrument types

Desired Outcomes: - Minimize manual effort to maintain investment tracking data - Increase completeness of counterparty exposure view (include MMCs alongside term deposits) - Reduce the frequency of workarounds for missing investment data types

Domain Insights

  • ON uses Anaplan for budgeting — potential data source for capex scenarios
  • MMCs at UBS behave like funds: invested amount changes monthly, interest reinvested as principal
  • ON is expanding rapidly (new stores, new countries) — creates unique forecasting challenge with no historical baselines
  • New Dynamics IT product manager is a champion for the BigQuery integration

Action Items

  • [ ] Jennifer/Giannis to publish investments dashboard to Rodrigo for testing
  • [ ] Amanda to share MMC/fund data (Google Sheet) with Palm team
  • [ ] Emma to schedule 30-min deep-dive scenario session with Lucía, Julia, and Amanda
  • [ ] Rodel to prepare integration specs/documentation for ON's new IT product manager
  • [ ] Emma to screenshot Amanda's MMC report for engineering reference

Notable Quotes

"Eight, nine. Super cool, actually. It's part of our goals, actually, for this year to do more scenario and stress testing." — Amanda/Rodrigo (on scenario prototype excitement)

"If the tool is not flexible or intuitive to make scenarios or adjustments, you end up doing it in Excel. And I think that's the worst that can happen to a TMS." — Lucía (from previous session 2024-11-19, referenced)

"I think I can see us using it for different reasons. Like, for example, for inventory. Like more this kind of networking capital." — Amanda/Rodrigo (on scenario use cases beyond treasury)

"Once you see your scenario, it would be nice to see what you said was your best case, your baseline, and your worst case, and then with the actuals laid on top of that so it doesn't wipe out the history." — Jennifer (on fixed scenario tracking)

Full Transcript

Meeting Title: On <> Palm Daily Standup Date: Feb 18 Meeting participants: Lucia Galan.caceres, Emma Sjöström, Rodrigo Cabrera, Amanda Mitt, Federico Morando, Jennifer, Giannis

Transcript:

Me: Agree. Things. I was like, no, I'm late. It was okay.
Them: If you are, compliment.
Me: Yeah.
Them: Your name?
Me: How's the coffee? Are you craving coffee?
Them: I made instant coffee.
Me: I would have done the same.
Them: It's not the same.
Me: No, for sure. But still, it's something, right? Jen.
Them: Hey. Tim.
Me: Rodrigo.
Them: Hey.
Me: Hey, Rodrigo.
Them: How is he going?
Me: Oh, good. Thank you.
Them: Very good.
Me: Hey. You.
Them: Doing great always. Snowing quite a lot here. I don't know if you can change my background. I changed my background to pretend that I'm in the. It looks real. Actually, no. I thought you were in the office.
Me: Yeah.
Them: I am in the office. But now I have a background. Nice. Good to know it works. Now. I need to know how to remove it. So what did you say? It's snowing. Yeah, quite a lot. I don't know how to remove a background. All signed off. Yeah. I don't know if you can see, but this super biggest snowstorm.
Me: All. Right.
Them: Surprised. Nice. Hopefully it's better next week, Giannis, that you're here. I don't know. I want to see some snow. Like I mentioned, this year was not my year with snow. Because it was knowing when I was in Greece, in the Netherlands, I missed it. So I hope to see some snow this year. You can always go to the mountains here. And I'm not sure you will see enough. You come for longer or just for the two days in the office in the city? Just for the three days. I had some plans already for the coming weekend, so I couldn't sense them. Otherwise I would stay longer. A bit. Next time. Next time. Are you saying you're just staying Wednesday, right? Or you're also staying Tuesday? As well. Otherwise I cannot make it to the Wednesday meeting. I'm coming to the office. Yes, I'm coming on Wednesday. Okay? Cool. You can come Tuesday if you want. Just in case you want to work somewhere not in night hotels. I might take you up on that. Let me see how the flight goes in the morning, because I think I arrived somewhere in lunchtime. And I will text. If I can make it easily with the hotel and stuff, I might drop in to the office as well. Are you staying nearby? I found an Nibis Hotel somewhere in the west, I think. Yeah. We are also into request. I think it's quite close. It's fine. Yeah, no worries. I'm a bit familiar with Zurich as well. Have couple friends there. Take the day off. And meet some friends. Yeah, unfortunately, all of my friends are back in Greece. That period, it's unlocking unlucky. You're stuck with Austin. That's also fine. Are we waiting for anyone else? No, we can start. Okay, perfect. Okay? Svesta. So, Rodrigo, I want to share with you quickly. The so since the investment data structure is now in place, I have been working in the back end to recreate the dashboard that you shared in the PDF that we discussed before, and it was actually relatively easy. So quite a lot of it is already now built. There is a couple of things, such as the weighted average investment rate that the team is working on refining, and I'll show you that in just a sec. But I hope that you might be able to decommission that mammoth sheet pretty soon. So I'VE got in here. A lot of the metrics that you have in yours. Obviously, one of the benefits is we can add filters into the page, too. A number of active investments are invested average term and I can say, well, I've weighted average rate in there counterparty spread. In both the pie chart and in a graph. Upcoming maturities. Interest earned by counterpart. I know that you have the exception for ups, and you can see it quite clearly here. This is a plot of all of the different. Trades that have been booked. I think it's. Too. Maturity date. So whether you like to fix it to deposit date or mature to date or both, we can do that. Average interest rate. This is the one where I've asked the team to refine this a little bit, because on the days where there's no deposit, It doesn't join the dots, so it goes to zero. So it looks a bit funny. And that also is why the average, when you total it over the period, is a little bit off. And then we can also see the interest rate development over time. So pretty soon I'll be able to publish this to you also, if you would like, in the custom canvas. The self serve. What I can do is let me just open this. Up. I can give you access to this data set, too. So, since we've discussed it, And, you know, sort of the limitations with the average. You can see. So if I set up a new report. Jen investment. Let's talk a lot. Let's just do something easy. I can add investments in here too. You can see I've added the forecast, the forecast data that we've been discussing that Giannis has created. So again, you can start using that for dashboards, too. But I can also add a nice investment, dated. So you could very easily rebuild that dashboard that I built. To. So you can amend it? I can build it in here for you. Maybe that's a good way to go. So you can update from there. And, yeah, that's kind of the big thing I wanted to share. There will be a page and up at some point in a nice design where you can manage your investments. However, in the meantime, All the stuff is in there. So be good for me to have a play around and let us know if there's anything that you can see that you would like to add, any additional fields that you would like to track as well. Thank you. I'll take a look and play around with. Nice. I'll publish that, Dash. But then, since you know the limitations, me and Yanis can publish that. So then you can see it and test it in the meantime. Thank you. And then learning from the other sessions that we've done on forecast scenarios, I might pass over to Emma earlier to share. We've been running over massively discussing how to manage different forecast scenarios with the new prototype because I know that Emma C became for you to have a look at it and get your feedback and us to sort of bounce ideas around. So, Emma. Go for it.
Me: If you're done or if you feel like there's nothing more urgent from your side.
Them: Nothing.
Me: I just received Sarah's version of the prototype. We still. Guys, we. We have a simple, fast prototype in the app that we're hoping to be able to release actually, soon. As. As a prototype, it wouldn't be fully functional, but it might be something for you to play around with and provide additional feedback. The data is wonky, so I'm going to lead with that. So you'll. You'll see what I mean. But I'm looking for two types of main feedback from you guys. One being on the. On the ui, like the experience. Do you think it would be useful for you? And the second actually going beyond what we're looking at right now. What kind of. Needs. Do you really have that? We could, if we thought if we were thinking about it in a bit, like, larger grounder, bigger way that we could help solve for you guys. That might be a bit abstract, but. I'll dive into it. So what we've done is we've. For now, it doesn't include investments. It likely should. Correct me if wrong, but. As of this prototype, we focused mainly on your operational cash. So to speak. And it's very similar to your forecast overview page. You can select filters, you can view it. In this case, I've just selected one entity. And then I can quickly just create what we call an assumption. And now I have two scopes that I can play around with, one below. Being the graph and the table that I see on this page. And the other one being the scope of the assumption I'd like to create. We decided for now to go with just simple model, more sort of like what if approach, percentage based. So let's say, okay, well, this is the entity I'm interested in adding something for because I know that we're expecting like sales to go up for whatever reason. Right. And then I want to see how that influences my predicted balances. And this is unfortunately, where the data is very off. So I have to add, like, I think something really large to even show something here. But the idea is for you to be able to preview then at different levels. Maybe you want to, you know, just see what I. This would just be currency. But let's say you apply an assumption across entities and across different currencies. You could. You could easily, like, visualize what the impact would be looking forward across entities, currencies, regions and so on. That's the general idea. You could also apply it across different time periods. You'd have more options than these. But you could also add multiple assumptions in one go to the same scope, maybe. And if you want to save something, you. You might want to, like, name it. You might want to explain why you want to save this assumption. What is it about? And then you. You'll see it that it's applied. And you can, of course, go ahead and add more assumptions like, oh, actually, for all of these entities, We know something is going to happen with Headcount, so you might want to just. Your influence, your payroll, for example. I'm just going to stop there. For thoughts from you guys.
Them: No. Very cool. I think I can see us, like, using it for. Different reasons. Like, for example, for inventory. Like more this kind of networking capital. Like account receivable, accounts payables, inventory. Maybe we can do some tests. We don't really do this for our forecast yet, so it will be new. It's something that, yeah, I've seen in previous experiences, like, yeah, just. You change the percentage. It was exactly like this. You change the percentages of caching or cash out. So, yeah, it's very intuitive and, I think, very fitting for Treasury. Like scenario planning. But yeah, I'm thinking for intercompany, would it also be possible to include like non business categories?
Me: Yeah.
Them: Or.
Me: I'm sure we can. We just had to limit the scope of the prototype right now because it's more complex than meets the eye. But 100% I do think we should allow for intercompany and also investments depth. Like I think as we add these capabilities, they should be. You know, visible here. So yeah.
Them: You said it like a view. If I understand it, like, for example, like a view that you would save on top and you can keep it for and go back to it, okay?
Me: Yes. Exactly. So we've. We've gotten some differing feedback. Right? So maybe it's all like, everyone has such a different mental model. Unlike what a scenario is, what an assumption is like. For me, this, that I just step on now is like a quick what if? Like, what if this happens? What if that happens? I can at least get, like, a quick just, like, overview. But let's say you'd like to. Actually, I don't know what's, like, a big event for on if you're opening up a new store, is that something that treasury cares about or is it, like, A more bigger things. Like what? What would be something you'd like to model out in anticipation of, like, event X or.
Them: I think this quite impressive. I have some follow up questions. But yeah, for example. If we open a new store in Sweden, we will be expecting many more cash outs in Swedish Grona than expected. So it will be interesting to see these assumptions. I guess we'll have to base upon a certain number that we have to come up with. Quite nice.
Me: Amazing, because one idea. Now I'm biasing you guys, so apologies for that. But one idea was discussed in another session, was almost like, treating these, like little Lego pieces, like these little assumptions that you could construct your own little scenario, like Swedish store opening, and you could, like, combine different assumptions, and then that is your full picture. And then if you apply that whole scenario, you'll see, like, okay, that's what will happen to my cash then. And plan accordingly. Or just be aware. Is that something that would be.
Them: Yes. In. Are these scenarios based on actual sort, based on forecasts, or based on.
Me: Forecasts for now. What would. What would be the need from your side with the actuals?
Them: I guess we are already considered the actuals for your forecast, so it makes sense that. Forecasting day or adding to the forecast.
Me: 100.
Them: Of course. I think some. In that scenario in Sweden might be tricky, I guess, because we don't have any actual. So nothing will be forecasted. How would.
Me: True.
Them: That. Maybe, like, input. I don't know. For store in Netherlands. This was our total payout. Maybe. I don't know. A percentage of that is for Sweden. Or what? What could be the. I don't know. I'm just thinking what could be a good way when. When there are scenarios that haven't happened before, or at least they haven't happened on that specific entity or currency?
Me: So for a completely, like, would it be a silly idea to have, like. Okay, well, just. Make your own.
Them: Made up.
Me: Like, I always think you will be like first 100k revenue per week. Then we think it will be this like, just like a super. I don't know. It's not even worth our while. Or is that just a really nice to have is also, I guess, a nice.
Them: Yeah, I guess it's just a really nice to have. But they had done. I'm just talking, thinking out loud. Because I guess. For us. A lot of things are new, right? So we haven't done anything. Or we are growing and expanding a lot, so everything is new. So it's hard to actually forecast for something that you have never experience, right? I guess for other companies they are more, more, more used to. But for example, I don't know, an acquisition, we have never done it or even thinking about it, so we don't know. But once the time comes, if it comes. It will be interesting to see. Okay, we have. We can forecast all of these scenarios.
Me: True.
Them: With? I don't know. Made of imaginary data that we input? I don't know.
Me: Yeah.
Them: It's all just.
Me: Yeah. No, for sure. Like what? Would it. Would it be anything like fx? Would that be something that's relevant? Or any other areas that are a bit, like, more. More close at hand for you guys right now.
Them: I think FX is always relevant. But not sure if we can forecast for a random currency Mexico. But then the effects impact. We also forecast how the currency development happens. For? Not sure what you were thinking. Or like effects.
Me: Like, super naive. What could be done right or something like. Maybe across all your bank accounts. For, let's say. Receivable for some reason. You think that for all your like all bank accounts that has a receivable inflow in in Mexican fees it will increase. Let's say sorry. I have to do, like, a ridiculous high percentage to even see a difference. In the. Okay. Yeah, she won me. It's slow for now because this data is mocked, so. Oh, that doesn't make sense at all. But let's say, let's pretend it does. Then you'd be able to see just based off, like, impact on one currency, and it's not working, but. Imagine this blue line would have probably. You know, upped your balance, for example, due to more sales in that currency, for example. So it would be an indication that you could see, like, okay, This currency will go up. And you can see that in your table. I don't think we've gotten that far yet in the prototype, but that's being the idea. Then you could see, like. The impact essentially. Is 1. One way of thinking about it as well. And we are deliberately trying to keep this quite naive to begin with because it's fairly complex.
Them: Emma. I was thinking about before. And then when you went onto the currency. There. Highlighted it a bit more because this table as well is looking at actuals, and then it goes to forecast. And it is obviously, I think sometimes you think about the scenarios. You set your different scenarios. Say on the currency. Look at the currency. One is a bit harder. But say just you. Worst, Casey, best case, in your baseline forecast, you've got your three scenarios that you'd like to run. And you're tracking what an over talked about this a little bit before. What scenario you are moving again so say for on they're going with their best case scenario all the stores are opening they're all going well. But if the forecast is always re baseliding to the actuals, you can't necessarily compare with the baseline versus the best case. So I think there is some sort of case. I kind of want to ask the guys if they agree that there is a bit of a case to setting the scenario. And fixing it. And then seeing the actuals laid on top of it. Whereas here and then the other forecast page. Too. The forecast scenario starts from where the actuals ended. But in some respect, once you see your scenario, it would be nice to see. What you said was your best case, your baseline, and your worst case, and then with the actuals laid on top of that so it doesn't wipe out the history. I don't know if I explained that properly, but similarly for fx, if it's called one way versus another way versus the point when you set your scenario. Would be nice to see them layered on top of each other.
Me: Yeah, for sure, for sure. We haven't really gotten to the whole best case, worst case, baseline space yet. It's something that's on our mind. There are so many different approaches to even get to those scenarios. And one thing now I'm running ahead a bit. But one thing we're also thinking about, if helpful, is actually exposing some confidence bonds. To your forecast so that you can clearly see, essentially, let's say, okay, we're with 80% certainty, your balance will end up in between these two bonds, for example, And our aim is to always make those bonds as slim as possible or like it will be. Obviously, the, you know, further ahead you're looking, it's going to be more like, less secure or a broader bond. So we're still experimenting a bit with that. But that could be something that might be helpful once we get to a place with scenarios that is about these kind of best case, worst case. But right now we are literally at the level of just Would it be helpful to even introduce something like this, which is more like a quick what if one off assumptions possibly being able to construct. Maybe scenario is the bad word, but little like, yeah, events or like buying a company or. I get that. The open a new store in a country you're not operating in right now is really tricky, but those kind of stuff. And then see how that might be applied across your best and worst case. Well, eventually.
Them: What do you guys effort? So you were opening Swedish store. Would you ever add that stuff? Add in full custard to think.
Me: Yeah.
Them: And then piece for variance analysis too. And a point of this scenarios. Yes, I guess that will be reflected at some point in the open payables, right? So eventually it will end up. Here. A forecast. The bigquery integration with the API. Oh, yeah, I have some updates. On that.
Me: Nice.
Them: So, yeah, we have a new IT product manager for Dynamics. So, yeah, he's, like, very excited about the project and about pushing AI into on. So I, like, spoke a lot about Palm with him and he checked our, like, business case and everything. And he just wanted to ask, like, if you have. If you have done this kind of integration with any other customers or, like, what exactly do you need? Any sort of specs, documentation. Yeah, so we're ready to kick off. But, yeah, he said, like, if you could share something beforehand so he can prepare a bit and see who to involve. Yeah, we are ready. He just said, like, yeah, we might get some resources a bit later on the year, like on Q2. But I think. I don't know. I think it's something we can. Like, he has no idea how much effort it will take. Right. For us or for you. So I think that's what. What he means just so he can plan who to involve, and then, yeah, if we would use APIs, how we would, you know, move on, like, ideally from your side, so, yeah. Which is good. Good news.
Me: Yes.
Them: Indeed. Very good news, Amanda. So I think the best way to go forward with this is let me pick it up with Rodel to ask him if there are some questions we can already send to you. But if we see that, yeah, the easiest is to just set up a meeting and have a discussion with the IT monitor, then we can also go with that and then find out all the details in that meeting. Yeah, it's just in case you have anything. Or if you don't. Sounds good. Thanks. Thanks. Yeah. So the Capex, like example for Sweden, I think would come from AP data, but we also have anaplan, so maybe we can create some sort of template. I was thinking for the Capex, uploads manual uploads into Palm. Right. I think that could work. For some cases because we do the reviews with the regions now more frequently, so maybe we can think of something like that in the meantime. Or for these cases. So we just. Yeah, we do some sort of data pool just for Capex. Yeah, I think that's good. So it can be so consistent on the budget. I think this opportunity, I don't know if you agree with this, but that on a plan budgeting, to feed more of that and to promise, you know, if you think that's correct, you know, because the more data points there are, the more accurate the forecast can be. I don't know if you ever wanted to go into that in more detail, but, yeah, always happy to help as well. Yeah. I think, for example, the categories we could maybe leverage a bit more. Yeah. Yes, we can start. Very cool.
Me: I'm just going to.
Them: No, sorry. Then it's a negative like we think WIX type account receivable to drop. Then you just add the negative percentage.
Me: Yes, exactly.
Them: Sorry. Okay.
Me: But, like, just, like, you can be honest, like, on a scale 1 to 10, how excited are you about this? This one feature here? Be. Be brutally honest. It's fine.
Them: I think.
Me: Oh, wow.
Them: Eight, nine. Super cool, actually. It's part of our goals, actually, for this year to do more scenario and stress testing.
Me: All. Right.
Them: At least. So I'm excited.
Me: Do you think it would be wild and taking a bit of. Let us have a think and iterate a little bit and then what do you think it would be while having like a separate, just 30 minute session again very soon, just like diving a little bit deeper into. What would be the core value out of something like this for you guys that you're hoping to achieve using a scenario tool? Just so we're making sure you're getting your job done the way you need it to get done. As well. Is that fair? Or a good idea.
Them: Yeah, sounds good. But I think, yeah, we should involve Lucia and Julia for sure.
Me: Yeah, that would be amazing, of course, to have them there as well. All right. Let's organize something then. That would be really fun. And cool. Thank you so much. This was really good. Initial. Initial validation.
Them: In the follow ups. Were there any bits from your side, Amanda, this week? Yeah, just one small bit. I just realized, like, we were so focused on the deposits, right? But actually they make way more sense for our like day to day cash management and everything. But we do have this product, the EMMC's, that are kind of similar to a fund, a money market fund in a way that like we just have this invested amount, it doesn't really change. So I think we might need to add that to the total investments if we want to have really the total investments in the total counterparty exposure. Because there with ubs. So we had. This is like. I think I can schedule a report like this. But I cannot add it to the ftds one, unfortunately, because there's no material. It's just like, yeah.
Me: Hey, Amanda, is it okay if I take a screenshot of this and share it with an engineer, or is it like, okay.
Them: No, I'll go. They can also share with you.
Me: Thank you. Just thank you.
Them: Yeah. So it would be a different Google sheet. Thank you. Yeah, we'll get that. I didn't do. Because you need the complete cash view for it to be helpful. I appreciate it. Otherwise, there'll always be Emmanuel workaround, and we don't really update it. Like, there's nothing automated with the bank, so the amount changes basically just monthly. Because we do an update in Caer, so it would be like once a month, because we don't really move. Like, we don't change that much, we don't really invest more, or like, withdraw from the fund in case that changes. Yes, we can discuss a new frequency, but I think for now, like, monthly. Could be good. Just a monthly, you know, refresh. Yeah, yeah, no, that. That makes sense. And does the amount just change? Is the interest reinvested? Okay, that is something we can easily model. So. Yeah, so the interest from the previous month becomes the principal. And then we can then calculate the future interest. That should be easy enough to do. We're a little bit over, but yeah. Thank you for your time today. And we will schedule that follow up session with Emma for the mortgage guys.
Me: Thank you so much.
Them: Thank you. Thank you. Have a good one. Thank you. Amanda, can you stay a bit?