This is the plan for October through December, written in September because the decisions it affects are being made now. Most of it is steadiness rather than news. The private beta opens to more firms, the same four jobs get better rather than joined by a fifth, and nothing new is scheduled to land in January.
- Our quarter is governed by somebody else’s calendar. It ends when the organizer replies start arriving, not on December 31.
- The beta opens to more firms, and most of that recruiting happens in November rather than spread evenly across three months.
- The scope stays at four jobs. Doing those four properly is the plan, and a fifth would be paid for out of somebody’s February.
- There are no ship dates below, deliberately. A firm choosing software in the fall needs to know what will be steady, not what has been promised.
- One thing we are not adding is a row of certification badges. A badge goes up when there is a held certification and a dated document behind it, and not before.
The quarter that ends on February 1
Fiscal quarters are a convention we do not really get to use. The firms Point was built for are accounting and tax practices, and their year has one date in it that does not move for anybody. The three months ahead are arranged around that date rather than around December 31.
The arithmetic is short. Organizer runs go out in December. The replies land in January. The IRS opens somewhere in the last week of that month, and by the first days of February the payroll filings and the broker statements are arriving on their own schedule from senders no human reads. A practice still learning a new tool at that point has picked the worst four weeks of its year to do it. What that means for a firm’s own planning is getting ready for the 2027 tax season, which takes the fall apart from the firm’s side.
What it means for us is a rule rather than a preference. Anything we intend a firm to be relying on in April has to be in front of them by early December, and after that the useful thing to be is boring. Software that changes shape in January cannot be leaned on in March, however good the change is on its own terms.
So this post commits to a scope and a direction and carries no ship dates. That is not modesty. A date missed by three weeks in an ordinary month is a date missed inside somebody’s busiest one, and the cost of that lands on a practice rather than on us.
Opening the beta to more firms
The beta is still private, and still per seat. What changes this quarter is how many firms are inside it, and the way in is the waitlist on the home page.
The recruiting is concentrated in November rather than spread evenly, and the reason is capacity rather than strategy. Settling a firm properly takes attention from real people at our end, and the number we can settle before February is not unlimited. Front-loading it means a firm that says yes has December to be new in, which is the month when being new is cheapest. Why that is true from a practice’s side, and what starting in February costs instead, belongs to the seasonal guide linked above rather than to this post.
Widening does not change the product for the firms already on it. Nobody’s setup is migrated, nothing is repackaged, and no capability is moved behind a different door. What it changes is how many practices are meeting the same product for the first time, which is where the useful complaints come from.
What the first firms actually ran into, including the parts we had wrong, is lessons from the beta. Who is joining, on what terms, and what the first week is really like is opening up to more firms.
Four jobs rather than a fifth
The beta is scoped to four jobs. Not because Point does four things, but because four is what we are holding ourselves to proving inside one season.
What needs you, at the top. The pile is ordered before you open it, so a morning opens on the two or three things that actually need a person.
An ask, turned into a dated task. When a client asks for something four paragraphs into a reply, it lands on your list carrying the date they named, without anyone retyping it.
Hostile mail held at the door. Risky messages are set aside before they reach you, and mail from a sender Point does not recognize is held rather than read into anything, so a stranger cannot give your inbox instructions.
A channel you can lock. Lock a message end to end and only the recipient can open it, Point included. That is a channel you turn on for a message rather than a property of your whole mailbox, and the difference is worth being precise about before anybody writes it into a firm policy.
Depth on those four means something unglamorous. The ranking being right about your clients rather than about clients in general. The date on the task being the date the client actually named. The quarantine catching a fake invoice without eating a real one. The locked channel being usable by somebody who has never opened one before, because a secure channel the other end will not open is a feature only on our side.
The rest of what Point does is real and in daily use, and it is not what the beta is measured on this season. The whole inventory is on the benefits page.
The screens Point already runs on
Point runs as a layer over the mailbox a firm already keeps, on Google or on Microsoft. There is no new address and nothing to migrate. Past the browser there is a desktop app of its own, the same feed on a phone where one swipe files a thread, and a list where attachments gather instead of staying buried in the threads they arrived in.
Those exist today. What is missing is not the surface. It is that somebody evaluating Point in October cannot see any of it from outside, and a capability nobody knows about is worth roughly what a missing one is. Three posts this quarter deal with that, one each: the desktop app, Point on your phone, and working with documents.
Writing down what already works is most of what a private beta owes the people considering it.
Putting the data answer in writing
The question that decides whether a firm can use Point at all is not a feature question. It is what happens to a client’s mail, who can see it, how long any copy of it lives, and whether the engagement letter now has to say something it did not say last year. That question carries a December deadline, because the letters go out in December and what they say is then fixed for the season.
So one of this quarter’s pieces of work is not code. It is setting our own answer down in the vocabulary a practice actually uses, which is how we think about data. The documents underneath it are already public and already dated: privacy and subprocessors. How to read an answer like ours, including the places where the honest version has an adjective carrying most of the load, is the questions to ask an AI email vendor. What a letter can and cannot carry is what to put in your engagement letter about AI, and it is a November decision rather than a December one.
One item is deliberately absent. We are not putting a row of compliance badges on the site. A badge goes up when the certification behind it is held and there is a dated document to hand over, and an unearned one is a false claim rather than a design choice.
What we are not building
A roadmap is more useful when it names what it declines, so here is that half of it.
Point is not a practice management system. There is no job list in Point, no view of how much work is already stacked in front of a preparer, and no software here will tell you which returns are not going to be finished this season. That judgment stays with whoever is running the season.
Point is not a document register either. If what a firm wants is a board showing which of eleven items have arrived for each client, that job belongs on a single list kept somewhere else, and why a document register belongs on one list sets out why the mailbox is a poor home for it.
Point is not a CRM, and not a shared inbox in the sense of several people working one queue together. A team joins with a seat each, and the surface they share is the business rather than the mailbox.
There is also no pricing announcement in this quarter. Private beta, per seat. Anything more specific than that would be invented, so it is not here.
The first three are all reasonable things to want. If one of them is the actual problem, the honest answer is that an email client is the wrong shape for it, and no amount of buying one will make it the right shape.
What this changes if you are deciding
Three situations, and they want different things from this post.
If your firm is already in the beta, the useful sentence is the one at the top: nothing you have started to depend on is scheduled to move between now and February. How far Point goes on each kind of work is still your setting and still sits where you left it, with every kind of work starting on review until you raise it yourself. That control is the autonomy dial, and what Point did on its own is listed with a timestamp in the activity log. Raising a setting for the season is a decision worth making in December rather than in March.
If you are on the waitlist, November is the month the quarter is built around. Nothing bad happens if you join in January instead, beyond the fact that you would be learning something new during the four weeks least able to absorb it.
If you are still deciding whether to change anything at all this year, this post should not be the thing that moves you. Its only real contribution to that decision is a negative one: there is nothing in the next three months worth holding out for. Whatever the answer is in October, the plan above will not have changed it by January. What does change when a mailbox is connected, and what stays exactly as it was, is how to switch to an AI email client, and the list is shorter than most people expect.
Common questions
Will anything change in the middle of the season?
Not by our choosing. The product a firm settles into in December is the one we intend them to carry into April, and nothing in this plan is scheduled to land in January or February. The part that does keep moving is the part that is supposed to, which is Point’s read of your practice as more of your traffic goes through. That is the ranking getting better at your clients rather than the software changing shape underneath you.
Why does a roadmap post carry no dates?
Because a date is the first thing in a roadmap to rot, and this particular audience is the one it rots on hardest. A firm choosing in October is not really buying October. It is buying the last week of March, and a promise that slips a little arrives inside the weeks with no slack in them. So this post names a scope we can hold to and leaves out the week we might miss.
Does opening the beta wider slow things down for firms already on it?
The product does not change for anyone already using it. What widening actually costs is our attention, which is exactly why the recruiting is concentrated in November rather than trickled across the quarter. A firm brought on in a week when we are paying attention is settled by December. One brought on in February is being settled in the month with no room in it.
Point does far more than four things. What are the other ones doing?
Shipping, and in daily use. The four are what the beta holds itself to proving inside one season, which is a different question from what the product contains. Scheduling, briefings, summaries, standing watches and the rest are all live, and the full list is on the benefits page. If your reason for looking is one of those rather than one of the four, that is a perfectly good reason to look. It is just not the thing this quarter is being measured on.
Is any of this going to change what we pay?
No number is being announced here, and none is being planned as part of this quarter’s work. Point is in private beta and priced per seat, and the plan above is about capacity and depth rather than packaging. Any tier, trial length or discount you see attached to Point did not come from us.
The short version
October through December is a capacity and depth quarter rather than a feature one, and the choice is about somebody else’s calendar rather than ours. The beta opens to more firms, with the recruiting concentrated in November so a firm that joins still has December to be new in. The scope stays at four jobs: what needs you first, an ask turned into a dated task, hostile mail held at the door, and a channel you can lock end to end. The surfaces that already exist get written down rather than replaced. Our answer on client data goes into plain language before engagement letters go out. Nothing is scheduled to land in January, no ship dates are promised here, and there is no pricing news. If that reads as a thin quarter, it is the right kind of thin. The firms this is built for are walking into four months where the most valuable thing a supplier can do is hold still.