In this Straw I set out details of the valuation I posted last night. I know Monday's webinar may well quickly date what I write here. However, it is a line in the sand because it sets out the basis for why I continue to HOLD $BOT. And that was a decision I needed to take today. If Monday brings new information, so be it.

SUMMARY

This Straw presents a valuation analysis of Botanix Pharmaceuticals ($BOT) based on the first six months of SOFDRA sales. Using updated data from the 8th July 2025 company webinar (The “Nightmare on Hyperhidrosis Street”), I built a scenario-based revenue model projecting to FY28 and applied forward P/E multiples to derive a valuation discounted to FY25.

Key components of the model include:

• Market Penetration: Three uptake scenarios (70%, 85%, 100%) based on prescribing dermatologist adoption over 24 months, with consideration of potential GP involvement and expansion of the specialist base over time.
• Refill Dynamics: Volume driven by average new scripts per prescriber and monthly patient churn. Churn is conservatively set at 18.5%, with sensitivities at 16% and 20%.
• Gross-to-Net Revenue: Gross sales per refill is AUD 1,500. A gross-to-net (GTN) ratio of 32% is assumed on average, reflecting early-year deductible effects and expected improvements in reimbursement management.
• Cost Structure:
• Sales & Marketing: Based on sales force headcount and dermatology industry benchmarks (AUD 462k per rep), with sensitivities up to AUD 20 million additional spend.
• COGS: Assumed at 7% of gross refill revenue.
• Other Expenses: Adjusted from the FY25 interim results, net of sales and marketing.

Twelve scenarios are modelled, combining varying assumptions for prescriber activity, churn, GTN, and S&M cost. The base case (P/E = 25) yields a valuation range of $0.22–$0.90, with a central (p50) estimate of $0.35. Even in worst-case scenarios, valuations remain above the current share price of $0.16- $0.18.

Using an alternative methd of applying an M&A revenue multiple of 5x FY28 revenues and discounting back yields valuations of $0.24 to $0.42. (Average is $0.33)

I conclude that despite recent market pessimism, SOFDRA retains strong risk-reward potential, and management has a reasonable timeframe to demonstrate longer-term value generation through platform exploitation and licensing.

INTRODUCTION

My basic approach is to model scenarios for revenue to the end of FY28, estimate the NPAT at that stage and apply a range of P/E ratios at that point, discounting back to end of FY25.

Revenues are driven off modelling total refills per month, using the detailed monthly history provided in the “The Webinar” (aka “Nightmare on Hyperhidrosis Street”, 8^th July).

The structure of the analysis is as follows:

1. The Revenue Model

1.1 Market Penetration

1.2 Refill Volumes

1.3 Average Number of Scripts Per Subscriber Per Month

1.4 Patient Churn

1.5 Gross to Net

1.6 So What Revenue Do I Expect

2. The Rest of the Financials – A “Ball Park” Estimate

2.1 Sales & Marketing Expense

2.2 COGS

2.3 Expenses

2.4 Getting to NPAT and EPS

3.     Valuation          

4.     Model Outputs Discussion

4.1 Discussion

4.2 M&A Valuation

5.     “A Nightmare on Hyperhidrosis Street 2 – The Revenge of the Applicator”

6.     So, What About My Thesis?

------------------------------------------------------------------------------------------------------------

1. The Revenue Model

1.1       Market Penetration

The market is large with some 3.7m seeking treatment in a Dermatologists office out of an estimated market potential of 10m.

The key volume drivers are therefore:

·      How many dermatologists (Derms.) are prescribing Sofdra

·      Number of new scripts written per month

·      How many refills each patient gets

We know there are around 4,000-5,000 Derms. who see patients with PAHh and will therefore assume 4,500 as 100% of the prescribing base.

The market penetration scenario assumptions are:

·      Maximum penetration achieved over 24 months

·      Penetrations of 70%, 85% and 100% modelled.

This is justified because very rapid penetration (51%) was achieved inless than 6 months. However, ultimate penetrations of 70%, 85% and 100% might at first glance appear unreasonably high. However, there are three further factors to consider:

First, the actual Derm base is 10,000-12,000, so if the product gains market acceptance, there is the possibility that the specialist prescribing base expands.

Second, the experience for the other anticholinergic in the market (Qbrexa) is that over time, some GPs will prescribe refills, or potentially write a script for a patient who has tried the drug but them come off it (for example, at first they couldn’t get the health fund to pay). Apparently, this has been written as acceptable by some health funds (Note: verification of this is required.)

Finally, the upside case (100% of 4,500) also allows for the potential that there are actually 5,000 prescribing dermatologists to begin with.

In short, while 100% penetration is unlikely, there is the potential for the prescriber base to grow over time.

A peak in number of prescribers is assumed to occur in 24 months from launch. The three modelled uptake scenarios are shown below. These scenarios are consistent with the observed fact that in the US dermatology treatments tend to reach plateau sales in the 3rd year.

Exhibit 1: Modelled Prescriber Uptake Scenarios

1.2 Refill volumes

Scenarios are generated for the number of refills issued by month. The assumptions in this model are:

·      Existing patients obtain refills, subject to a Monthly Churn Rate (% Churn).

·      Active Monthly Prescibers write “n” Scripts per Month

From this simple model, the number of refills in any month is simply:

TRx(n) = Total (Re)fills issued in Month “n” =  TRx(n-1) . [1 - % Churn] + NRx(n)

where

NRx(n) = number of new scripts (i.e. new patients, including returning patients) written in the month.

In turn we can find NRx(n) from the Total Number of Prescribers (n) x # Scripts per Subscriber Per month.

So, we have two key variables we now have to understand:

·      Number of new Scripts written on average per Prescriber

·      % Monthly Churn.

I’ll next look at each of these in turn.

1.3  Average Number of Scripts Per Subscriber Per Month

Here we turn to the data from the first 6 months from The “Webinar”, and perform the analysis shown in Exhibit 2 below:

Exhibit 2: Model Calibration – New Scripts and Churn

Source: The figures in blue are from the “Webinar”.

I’ve estimated the New Scripts in each month (the Churn model is described in the next section). From this, we can calculate how many new scripts were written per Prescriber in each month.

Interestingly, the number started very high, which indicates that early prescribers might have already “warmed up” by having been engaged over the prior 3-6 months as part of the Patient Experience Program. In any event, $BOT presumably had a kernel of super-prescribers and KOLs ready to go at launch.

In previous Straws, we’ve also spoken on this forum about whether a potential “bolus effect” exists. The would be from highly motivated patients aware of the products approval and actively seeking it after launch.

So, the rapid fall-off in the Average Number of Scripts per Prescriber per Month is unsurprising. In fact, we expect it.

A source of error in this analysis is the Churn model leading to an estimate of the number of patient “Lapsing” each month. I’ve played around with different “% Churn" values, and the overall observation is robust.

Scripts per Prescriber per month falls rapidly over the first 6 months, although appear to be levelling off. This is reasonable if the initial population of "super-prescribers" gets diluted by the more general population and/or if the “bolus” effect dissipates rapidly in the early months after launch.

Now, the key question is how this number changes over time.

There is evidence from other drug launched in dermatology, that indicates that the prescribers initially prescribe at a low level, and that this grows by 2-3x over the next 12 months.

Whether this proves to be the case for SOFDRA is one of the big value drivers and uncertainties. At this stage it is unknown.

Note also that I am ignoring the prevalence of the condition at this stage. It doesn’t matter, because the results of the model represent a very low proportion of the prevalent population, so Sofdra will not be limited by the number of patients seeking treatment.

Conservatively, I have generated the following 3 scenarios, which I hold as independent to the number of prescribers:

Exhibit 3: Scenarios for New Scripts Per Subscriber Per Month

I have clearly excluded the scenario that the product “flops”, and clinicians reduce their prescribing over time. This scenario cannot be ruled out, and could occur under two situations.

·      Clinical Data doesn’t support continued use (for whatever reason)

·      A superior treatment emerges.

I’ve ruled out the first case because of the clinical trial data published by the JAAD, but also the experience in Japan, which showed consistent growth over the first three years in the market. While the product is only partially effective, it appears to be well tolerated and delivers a sufficient benefit to be meaningful to a reasonable proportion of patients who try it.

At this stage, it is unclear whether a superior product will emerge in the near future.

Note: This sensitivity most significantly impacts the valuation.

1.4 % Monthly Churn

The Webinar covered a lot of information about adherence and number of refills. I’ve developed a basic monthly churn model, simply because it is the easiest way to fit the data provided for the first 6 months, and then to project forward.

If 18.5% of patients churn off the drug each month and don’t return, we get the following profile:

·      3.46 total fills from February to June (i.e.2.46 refills)

·      Only 11% of patients remain at the end of the first year

·      4.94 fills over the first year

The first point fits the data presented by Howie in the webinar.

On the second point, no-one knows how many patients will come back for another script at the start of the second year. However, 11% seems a conservative approach. Perhaps more will if a significant proportion of reimbursed patients perceive value from the product. So, there is a significant potential upside that I have not considered, as I am choosing a cautious approach in absence of data.

The % Monthly Churn model is flawed. For example, we know a proportion of patients are not going for auto-refills, and are maybe only trying 1 or 2 refills, before abandoning the treatment. However, I am basically comfortable with the model as a rough estimate, given that:

1.     It predicts well the average number of refills for the February patients over a 5-month period

2.     It aligns with managements enthusiasm that the product is performing well above the norms for dermatological products, which have of an average of 2 fills per patient (i.e., only 1 refill).

I will run two sensitivities on this parameter at % Churn levels of 16% and 20%, noting that at a 20% monthly churn, only 9% of patients are still using the product in the 12th month after first prescription.

This is an area of high uncertainty, and based on performance over the first 6 months and management’s statement about the February Scripts, there is a possible material upside risk to this factor. Rather than introduce further model complexities, this will be something to revisit over time.

1.5 Gross-To-Net (GTN)

$BOT appear to be achieving Gross Sales of AUD1,500 per refill. So with modelling the volume of scripts well-defined, the next big parameter is GTN, in order to achieve net revenue. Management believe they will ultimately achieve a GTN of 30% to 40%.

The exit rate for June was 23%, improving at about 2% per month. Given that Q1 and Q2 are the high deductible season, recovery to the mid-range seems likely.

Secondly, I expect management to tighten the copay policy in Year 2, and also for optimisation of Pre-authorisation of Scripts over time to improve GTN over time.

Analysis from studies of other drugs in sectors like derm. shows that Q1 and Q2 are typically hit by the high-deductible period, with stable revenues in Q3 and Q4,

I have therefore derived the following assumptions based on other studies (note: at this stage $BOT management haven’t said much about this):

·      GTN continues to improve at 2% per month, reaching 35% by end of calendar 2025.

·      Thereafter, every year, there is a Q1 hit to 72.5% of the Q4 value, and in Q2 88% of the Q4 value, with full recovery by Q3.

We don’t yet know what the Q1 and Q2 annual deductible hits will be. However, the chosen values seem reasonable given experience elsewhere.

The net effect of the 35% assumption, and the annual resets lead to an average annual GTN of 32%. (Note: this is down from my original valuation of 50% - a bit hit to value!)

I have not run any scenarios or sensitivities on GTN. Who knows, perhaps average annual GTN is only 28% or maybe it is 36% - these are now relatively small uncertainties compared with others discussed here! So, I’ll settle with 32% as a reasonably conservative but not unduly pessimistic number. Exhibit 4 shows the GTN over time.

Exhibit 4           GTN over Time

This concludes the revenue model assumptions.

1.6 So What Revenues Do I Expect

According to my model, Sofdra will generate peak revenues of anywhere between $AUD137 and AU$240m by FY28 (or US$90m – US$160m).

That’s very materially down from upside cases I was projecting of anywhere from US$200m to US$600m only a few months ago. (Sad face emoji)

Of course, it is possible that in every assumption I’ve made in this model I’ve suffer from a negative bias induced by the “Nightmare …” and there are certainly upsides I’ve chosen not to consider, particularly around GTN optimisation and, more materially, increasing prescription rates over time.

But rather than “fudge” my model, I’ll run with what it's telling me and – if warranted over time – I'll make adjustments in the light of evidence.

2. THE REST OF THE FINANCIALS

With a high range of uncertainty around the revenue model, I have kept the rest of the financial modelling simple. I’ve also not spent any time trying to get a sensible number for FY25 simply because it is a transition year, with several non-recurring factors:

·      Platform build

·      Launch preparation

·      Onboarding of Sales and Marketing Staff

·      Launch inventory build

I want to emphasise this because I will not judge this model by how well it predicts the FY25 Full Year result. I’ve spent zero effort trying to do that because it has no bearing on the company value in the medium term - even though it may well drive the market.

The major uncertainty is the spend on Sales and Marketing. So my approach here, is to take the Expenses from the 1H FY25 Accounts, back out the Sales and Marketing element, and build a simple sales and marketing cost model.

 2.1 Sales and Marketing Expense

I will estimate the total Sales and Marketing Expense as follows:

S&M Expense = Sales Force Headcount x Benchmark value

This is a crude but well-established method in pharma to derive total S&M Expense from the size of the field force, with the benchmark picking up all related and overhead costs.

Reasonable benchmarks in Dermatology are anywhere from $USD 300 k per FTE to $500 k per FTE, which turns into AUD 462 k to AUD 770 k.

We know that $BOT have hired an “A” team of derma industry veterans. And “first product” businesses usually pay over the odds. This will be offset by the fact that some of the expenses covered by the benchmark are already “hidden” in other lines of the Accounts – given the AASB/IRFS model applied in Australia.

Therefore, the approach to be followed is as follows:

·      Assume AUD 462 k per FTE

·      Run high case sensitivities of AUD$10m, $15m, and $20m (for a 50 strong field force, these sensitivities are equivalent to AUD200k, 300k and 400k per head – so they should cover the potential outcomes).

These sensitivities are also important because we don’t know how much digital marketing spend has been thrown at the business.

Most of the platform build will be included in the 1H FY25 accounts, So we don’t need to worry about that. However, it is clear to me from the Webinar that $BOT are not yet spending big on digital marketing, and as I’ve written previously, that they are seeing the highest ROI on investing in the good old door-knocking salesforce.

Why is this the case? Well, it appears the physicians are easily “activated.” So rational resource allocation is to get your reps in front of all the 4,000-5,000 target derms. asap! Which is what management appear to be doing.

For the model, Sales and marketing is built up as follows:

·      FY25: 27 Reps (FY25 is not refined as it's immaterial to valuation)

·      FY26: 50 Reps

·      FY27: 60 Reps – they go from 90% coverage to expand the base in targeted areas.

And so the Sales & Marketing expense then follows.

2.2 COGS

I’ve assumed a flat 7% of Gross Refill Value is assumed. i.e., 0.07 x AUD1500 = AUD105 per refill

This is on the high side to allow for Tariff impacts (assuming Tarrifs apply to COGS and not Sales!)

An error in the model is that inventory needs to be made 3-6 months ahead of sales, but this is not material given all the other assumptions, so I’ve ignore working capital.

2.3 Total Expenses

Expenses are estimated as follows:

·      The expense base at 1H FY25 Accounts (4D) as starting point: AUD 32m x 2 = AUD 64m

·      Strip out Sales and Marketing, so it doesn’t get double-counted: -AUD 17.7m

·      Expense Base = AUD 47m + Sales and Marketing.

I’ve assumed interest is included in here, and may have under-estimated charges for the expanded debt facility.

2.4 Getting to NPAT and EPS

PBT = Net Revenue – COGS –  Expenses

Tax rate assumed at 25%, as there will be benefit from carried forward tax losses.

NPAT = PBT * (1-Tax)

Shares on Issue: Management are using a fair amount of share-based compensation, so I assume 3% dilution p,a,

3. VALUATION

The model generates FY28 NPAT for 12 scenarios, combining the various factors covered.

P/E Ratio – This is the second big driver for the change to my valuation. The change in this valuation over my pervious valuation is that SOFDRA does not appear likely to be a blockbuster. It looks like it will be moderately successful and reaches maturity in FY28, and is not rolled out beyond the US. (The economics are not attractive.)

Growth from FY28 onwards will then depend on whether – over the next three years (not tomorrow!) – management can bring other undervalued dermatology drugs onto the platform.

I’m not sure they’ll succeed and so the P/E ratio scenarios I will apply in FY28 are 20, 25 and 30.

If you think $BOT is “Sofrda and done”, then eventually it will get bought out at some multiple.

So, I’ve taken FY28 EPS and discounted back for 3 years to end of FY25 at10%

Bingo.

This is still betting on management experience and skill in dermatology, and it gives them a reasonable time horizon to either do platform deals or licence in new molecules. If I didn’t believe in management, then P/E scenarios of 15 and 20 would probably be more appropriate. This risk is not explicitly modelled, but that’s because I believe management will find a way to create more value over time.

The detailed inputs and key outputs are listed in Exhibit 5.

Exhibit 5: Model Scenarios and Outputs

The table above shows the outputs from the various scenarios. I’ve not really had the time to think about the scenarios probabilistically, but if I had, the distribution of valuations would be as shown in the Exhibit 6.

Exhibit 6: $BOT Valuation Results

At my refence P/E of 25, I get a valuation range using my usual p50% (p10% - p90%) notation, of $0.35 ($0.22 - $0.90) in roundabout terms.

At my p50% level, the range generated by my P/E values (20, 25, 30) are $0.27 to $0.41

My conclusion is that the market has indeed over-reacted to the “Nightmare on Hyperhidrosis Street”. Even in my lowest case analysis, I can’t get below $0.17. And yet that’s where we are today at $0.16 to $0.18.

While my previous very bullish view on $BOT has been materially deflated (sigh), I think the market has got this one wrong. Standing here today, you’d probably need to give me $0.60-$0.70 to get me to part with my shares.

4. Discussion of Valuation and Model Outcomes

4.1 Discussion

Depending on how you compose your scenarios, you can generate either more valuation results at the low end or more at the high end of Exhibit 6.

So, picking a number is indeed a fools game. I’m not sure of the value of doing more analysis on this, simply because the spread of valuations starts squarely at today’s market price and are solidly risked to the upside. IF YOU BELIEVE MY ASSUMPTIONS.

It is true that I could easily generate valuations down to $0.10 or lower, but equally, I can easily still get valuations north of $1.00 – in both cases using reasonable assumptions.

But based on what I believe to be reasonable assumptions, I am a solid HOLD on $BOT given by 4% RL position.

4.2 M&A Valuation

If we assume that by FY27 it becomes clear that $BOT is nothing other than “US Sofdra and Done”, then it won’t make sense as an ongoing entity and will get acquired.

To test the valuation, I’ll apply a modest 5 x FY28 revenues, and discount back.

Doing this I get a range of valuations of $0.24 to $0.42. Funnily enought, the midpoint of $0.33 is eerily close to my bottom-up $0.35 p50% at P/E = 25. (Honest, I haven't had time to fudge the models!)

5. “A Nightmare on Hyperhidrosis Street 2 – The Revenge of the Applicator”

So, why another “comprehensive” webinar on Monday?

I think management HAVE to do this because the 4C is doing to drop on Monday. Revenue and Cash will be bugger all, and cash burn will be scarey. And so management has to help the market make sense of the cost base. If they don’t do that, half the analysts will predict that $BOT runs out of money pretty soon.

And I don’t think they will run out of cash. For example, I’ve plotted the financials below for one of my more central case scenarios in Exhibit 7. $BOT can get close to breakeven in FY26 and is strongly cash generative in FY27.

Exhibit 7: Modelled $BOT Financials (Scenario 6)

Even in the case where I’ve layered on $20m of excess sales and marketing costs, with the lowest monthly prescription case (Scenario 12), there’s probably enough liquidity to get through to positive cash flow in FY27, just.

Exhibit 8: Modelled $BOT Financials inHigh Cash Burn / Lower Revenue Case (Scenario 12)

Of course, I’m also hoping for some more insights about rollout. After all, there’s been another 4 weeks of data, so hopefully there’ll be an update on scripts and prescribers.

6. So. What about My Thesis?

The whole point of doing all this work was find out if my investment thesis is intact or not. And?

My investment in $BOT was initially predicated on the view that the market was seriously mis-pricing development and execution risk. Development risk mispriced, because we knew the product works based on experience in Japan and the promising US clinical trial data (now published in JAAD). Execution risk overblown because 1) there is huge unmet need in the market, 2) the management team have a strong track record in dermatology launches and 3) the existing anticholinergic product in the market has well-defined deficiencies, and is being managed by a lightweight company trying to juggle multiple products.

I bought $BOT between $0.325 and $0.47 in the belief that this business was worth anywhere from $1.00 to $2.00.

Wind forward to today, and while physicians are getting onboard, prescription rates are underwhelming, and conversion to net revenue is less than (I) expected. Added to that, the company is rapidly scaling up sales and marketing spending.

So, the market is in the doldrums, seeing this business as worth $0.16 - $0.18. But I think it would be worth anywhere from $0.20 to $0.90. That’s a serious haircut to what I thought, but still an interesting investment. And it is early days.

So, my thesis while seriously diminished, is not broken. I’m happy to see how this story continues to unfold.

At this stage, I’m not sweating.

Disc: Held in RL(4%) and SM