The Boring Monopoly That Governments Can't Uninstall

In March I wrote that Objective's government customers cannot realistically leave. The market spent the autumn agreeing with me rather too enthusiastically, running the price from A$11.50 all the way to A$23.10. Then on the first of July the Department of Defence, the company's flagship reference for 25 years and roughly 140,000 users, declined to renew its support agreement. The shares last closed at A$7.19, down 69% from that high. Around A$1.5 billion of market value evaporated in a matter of weeks.

I have always said I want to know where I'm going to die so I never go there. Well, here is the corpse, and my fingerprints are all over the map. The only respectable thing to do now is a proper autopsy, performed on myself, in public.

What Actually Happened

On 1 July 2026 Defence declined to renew the ECM Upgrade and Support Program. The immediate profit effect in FY26 is nil, which is the sort of fact that comforts people who read only the first sentence of announcements. The real damage sits one year out. The contract was worth somewhere between A$13 and 17 million of recurring revenue, and support revenue of that kind carries very little cost against it, so most of it was profit. My base case now has FY27 operating profit falling about 17%.

It gets worse before it gets better. The licence position is unresolved for every one of those 140,000 users. Objective's stated position is that entitlements above 85,000 users were tied to the support term that just expired. Defence, for its part, says it intends to keep using the software. So the company's proudest reference customer is now its largest open dispute. You could not script a nastier reversal.

*The market's opinion of one company at three moments inside five months. Note where the dotted line finally crosses my old buy target. Source: ASX market data, author's March narrative.*

An Uncomfortable Piece of Arithmetic

In March I set a target buy price of A$10.03 and said we wait. By June that looked like senility, with the stock at twice my target and the neighbours all rich. By mid July the market obliged me after all, at speed. Anyone who took that number literally and left a resting limit order in the market got filled on the way down and now sits on a 28% loss inside six weeks. The fellow who chased it at A$23.10 is down 69%, which is the only comfort I am entitled to, and a poor sort of comfort it is.

The target price rested on a fair value of A$13.38, and that fair value rested on an assumption that no large customer would ever walk. When the assumption died, the number attached to it died too, but a limit order does not know that. A price target without a thesis check attached to it is a loaded gun left on the kitchen table.

So the first lesson costs nothing and is worth the most: the discipline was never "buy at the number." The discipline is "buy at the number if the reasons still stand." Anyone who cannot tell those two apart should own index funds and enjoy their weekends.

*A 69% fall needs a 221% rise to get home. This is why rule number one is what it is. Source: arithmetic, which has no opinion.*

What I Got Wrong

Three things, in descending order of embarrassment.

First, I treated switching costs as a law of physics when they are merely a strong tendency. A determined customer with its own IT ambitions and a budget the size of a small country will eat years of disruption to prove a point. The moat argument was right in general and wrong in the particular, and portfolios are built entirely out of particulars.

Second, concentration. Defence was about 11% of recurring revenue. Neither the company nor I ever put that number in front of you, because the company never disclosed it and I never forced the estimate. When one customer is a ninth of your recurring revenue, "customers cannot leave" should have been written "one customer matters far too much." I scored management 4 out of 5 in March. The renewal negotiation ran from March; in February the market was being told of confidence in the outlook. The chairman and the chief executive are the same man, he controls 64.8% of the votes, and the governance raters at ISS have the company in their worst risk decile. Alignment is real. So is the fact that nobody outside the building knew the biggest contract was wobbling. And I note, without further comment, that no director has bought a single share since the crash.

Third, I let a good story do work that only numbers should do. "Toll bridge" is a lovely phrase. It is not a substitute for asking who the biggest truck is and what happens to toll revenue if it finds another road.

What Is Still True

Now invert. If the thesis were truly dead, what would the ledger look like? Not like this.

In the December half, revenue grew 9% and profit grew 10%. Planning and Building recurring revenue grew 29%, and every council that adopts the software drags its local developers and architects onto the platform with it. The Scottish Government committed to Nexus. There are still more than 2,000 public sector customers, 84% of revenue recurring, gross margins around 94%, A$95 million of cash, and not a dollar of external borrowings. The founder still owns nearly two thirds of the company, the share count has gone down rather than up for three straight years, and another 586,963 shares were cancelled in June. Outside the Defence contract, the engine was compounding at roughly 9%.

The bridge lost its biggest truck. It did not fall into the river.

The New Numbers

My valuation had to be rebuilt, and rebuilt more humbly. Quality tier drops a notch, because the tier system exists precisely to price the sort of event that just occurred. The scenarios now read:

Weighting those at 30/50/20 gives an expected value of A$7.92 against a price of A$7.19. Call it 10% of upside. Interesting. Not compelling.

For those keeping score across this series, my estimate of value has gone from A$17.68 last September to A$13.38 in March to A$8.27 today. Some of that was better method. Most of it was worse facts. Either way, a man whose value estimate roughly halves inside a year owes you smaller conclusions and bigger discounts, and I intend to pay that debt in full below.

*Every scenario moved down by roughly 40%. The price moved down more. Both facts matter. Source: author's DCF, March and August 2026.*

The more instructive exercise is to run the machine backwards and ask what today's price assumes. The answer: roughly 4.6% profit growth, forever, from the reduced base. Australian nominal GDP runs about 4.75%. The market has stopped asking this company to be special. It is asking it to merely exist.

*From ambition to resignation in five rows. The price now assumes less growth than the economy. Source: company filings, author's reverse DCF.*

That is a modest ask against a business still compounding at 9% outside its wound. It is a fair ask against a business that just demonstrated its worst case is not hypothetical. On 18.5 times trailing earnings, against TechnologyOne at roughly 78 times, the market has moved this stock from the compounder shelf to the annuity shelf. In March I called the multiple gap an opportunity. Today part of that gap is a verdict, fairly earned. Both readings are honest, which is what makes this hard.

What I Am Doing About It

Sitting on my hands, which is where most of the money in this business gets made anyway.

The verdict is Hold. The price sits about 13% below my new base value, and for a company that just showed me its worst case in the flesh I want a 35% discount, not 13. That puts the new buy target at A$5.38. Yes, I am aware of how it looks to cut a buy target from A$10.03 to A$5.38 inside five months. It looks like a man who was wrong. He was. The target follows the facts, and the facts got worse.

The diary entry that matters is Thursday 27 August, when FY26 results land. I want to see closing recurring revenue near A$120 million, growth outside Defence holding at 9% or better, and some honest words about the licence dispute and about how much development spend is being parked on the balance sheet rather than run through the profit line.

My exits and entries are written down in advance, since I have just demonstrated what happens when they are not. I walk away entirely if a second major customer severs, if closing recurring revenue prints below about A$118 million, or if margin guidance drops below 35%. I buy if the price reaches A$5.38 with the thesis intact, if a Defence settlement recovers more than half the lost revenue, or if results show 10% growth outside Defence with decent FY27 guidance. In the meantime the company pays 26 cents a year, unfranked, about 3.6% at this price, which is modest rent on my patience.

In March I closed by saying there are no called strikes in this game. Still true. But the last five months added a clause: there are also no called strikes for the pitcher, and this one just proved he occasionally throws at your head. Same strike zone. Better helmet.

The big money was never in the buying or the selling. It is in the waiting. And the waiting just got a great deal cheaper.

---

DISCLAIMER: Based on information available as of August 13, 2026. Not financial advice. All data must be independently verified before any investment decision.