So, What's
Actually Next?

We get asked what's next often enough that we would rather answer it in one honest place than scatter half-answers across a dozen conversations. Six things are actually on the table right now, at six genuinely different stages. Some are running today behind the scenes. One is barely more than a question mark internally. We would rather show you exactly which is which than describe all six the same confident way.

1. Routing Engine 2.0

Today's routing engine builds its answer fresh every time someone requests a route: it grabs the coastline data for that stretch of sea, scatters stepping stones across it, and finds a path through them in a few seconds. It is genuinely good at that job, checking real depth, learning real shortcuts through rivers and channels, and it has already had every disproportionate-detour bug we could find tracked down and fixed with real before-and-after numbers.

Routing Engine 2.0 works differently underneath. Instead of scattering stepping stones fresh every time, it precomputes a huge map of the sea once, tens of thousands of water cells connected by hundreds of thousands of possible paths between them, packed tightly near straits and hazards, spread wider in open water. A route stops being something calculated from scratch and becomes something looked up. That should mean smoother paths, faster answers, and a system that gets better the more it is used instead of paying the same cost every single time.

Right now, that is running on a private test page only our own team can see. It is not yet the thing planning real shipments, and it will not be until it has been proven the same way the current engine was: real depth checks, real edge cases, real regression testing against actual coastlines, not a demo that only has to look convincing once. We wrote the full engineering account of this one already, including exactly what today's engine still cannot do that 2.0 is being built to fix.

2. Apple ecosystem

The staff app and the customer app both exist today, both work, and both are Android. That was never a snub toward Apple, it was the honest first move: Android is what most crews on deck and dock staff already carry, what a fleet already knows how to provision in bulk, and what costs less to replace when a handset gets dropped in salt water, which happens more often on a working vessel than in a normal office.

Apple ecosystem support means the same two apps, the same functionality, reaching iOS. Not a redesigned product, not a different feature set, just the same job running on a device family it does not run on yet. It is already flagged as coming inside the platform itself, not a promise invented for a blog post. What we do not have yet is a firm date, and we do not have fleet device management worked out for operators who would run this on company-managed iPads rather than personal phones, which is a real, unresolved detail for any operation planning around it. Android stays fully supported the entire time this is being built. It is not a placeholder waiting to be replaced by something better.

3. A subscription plan

All-in-One being a one-time payment is not an accident we were quietly reconsidering, it is a promise we still stand behind, and it has not changed. What we heard was a real, recurring request from smaller operators who wanted to start lighter than a full installation. That is now live: three hosted tiers, billed yearly, with a free trial and no card to begin. Nobody who bought All-in-One pays a penny more than they did, and nobody has to move.

Update: this shipped. See pricing for the tiers, or build your own if you only want part of it.

The honest reason it is not shipped yet is that it is harder than picking a monthly number. Domain-bound licensing, business provisioning, and six months of included maintenance were all built around a one-time buyer. A subscription has to make sense against that same infrastructure without quietly becoming the exact kind of recurring bill we built this platform specifically to let operators avoid. Getting that wrong in either direction, either underpricing it into a support burden or overpricing it into a worse deal than the thing it is meant to make more accessible, is worse than taking longer to get it right. We wrote about exactly what is stopping us in more detail already.

4. A hosting service

Self-hosted has never meant free to run. Every client today already pays for their own server, somewhere, to someone, a cloud provider or their own data center, and that cost exists whether or not we ever mention it. We now offer to run that infrastructure for you instead, for operators who would rather not hire for a skill they do not otherwise need, or who do not want to be the ones woken up at 3am when a disk fills up.

Update: this shipped, in two shapes. The hosted tiers include it. If you own an All-in-One licence and simply have nowhere to put it, we will run it on AWS, Google Cloud or a hosting partner, in your account or ours. It stays your licence and your database; cancel and you take the server with you.

The software itself stays a one-time purchase either way. Hosting would be billed as what it honestly is, real server, bandwidth, and operations cost, the same category of expense you would pay a cloud provider directly, not a software subscription wearing a different name. We are still working through pricing, which regions we can realistically support well, and what a service-level commitment looks like, including whether data residency inside Indonesia is offered as a default or a choice, given how much of our client base operates under exactly that expectation. None of that is locked yet. Self-hosting is not going away and stays the default. This becomes an option next to it.

5. Hardware

Routing and tracking are only as good as the data actually reaching the platform, and that is where the honest gap sits today. Some vessels run on GPS and AIS equipment that is old, inconsistent, or reports infrequently, and once a vessel is genuinely at sea, connectivity to send any of that data back can disappear entirely for long stretches. Bathymetry is a good example of the limit: our routing already checks real depth against datasets like GEBCO, which we have written about in detail, but that is a static snapshot of the seafloor, not a live reading of what is actually under a specific keel on a specific day.

Hardware is how we are exploring closing that gap, not replacing the software, adding to what feeds it. Sonar-style depth sensors that report real-time water depth instead of relying only on static charts. Dedicated marine GPS and AIS units, built for the job the way a Garmin-style unit is, rather than a phone doing its best. Other sensors worth a live feed: engine hours, fuel, whatever an operator's specific fleet actually needs eyes on. None of this is designed, sourced, or committed to specific partners yet. This is genuinely the earliest-stage item on this entire list.

Connectivity is its own honest problem, separate from the sensors themselves. A depth sensor or a GPS unit is only useful if its reading can actually reach shore while the vessel is still at sea, and for a lot of Indonesian waters, reliable connectivity offshore is not something any single company builds alone. The current thinking leans toward partnering with an existing satellite or internet provider, the kind of thing Starlink and similar services already do, rather than SEALUTION building its own network. But which provider, on what commercial terms, and for which parts of our client base first, is still genuinely unresolved. We would rather say plainly that this part is unsolved than describe a partnership that does not exist yet. The platform already works fully without any of this. Hardware would only make the data feeding it richer and more current, not replace what already runs today.

We are documenting the deeper thinking as we go, the mounting and safety questions, how an install would actually run, the connectivity tradeoffs, unfiltered and including the parts we have not solved.

6. An IPO?

People occasionally ask, so here is the honest, deliberately boring answer: it sits somewhere on the very long list of paths a growing company thinks about, the same way most do. Nothing is decided, there is no timeline, no target exchange, and no valuation conversation happening. Nothing in this section is investment advice, and nothing here should be read as an offer, a solicitation, or a promise of any kind regarding shares in the company. We are noting that the question exists, not announcing a plan. If that ever changes into something real, it will get its own honest post when there is actually something to say, and this section will not have gotten any longer between now and then just because a blog post asked for more detail.

Where this stands

One of these is a working prototype we can see today, just not one we trust with a real shipment yet. One is scoped and in progress, waiting on details rather than direction. Two have since shipped, and this page has been updated to say so rather than left to age. One is genuinely early, more question than plan. One is barely a plan at all, and we are comfortable leaving it that way in public until it is something more. The roadmap is only useful to you if it tells you the truth about where each piece actually stands, so that is what this page is for.