Most firms have had some version of the same conversation with a software salesperson. You ask which states the package supports and you get a number. Forty states. Every state with an income tax. The number is usually accurate in its own way, and it is almost never an answer to the question you were actually asking, which is whether your clients’ returns will go out electronically in February.
I have spent the past several months on the other side of that conversation. I am an enrolled agent with a one-person practice, and I am also building tax software for it, which means applying to state and territory developer programs one at a time. Over a hundred regulators. What I learned is not that vendors are dishonest. It is that state approval is more fragmented than any sales conversation can convey, and that a few specific questions will tell your firm more than a coverage map ever will.
Approval is per state, per tax type, per year. There is no central clearing house. Each state runs its own vendor registration, issues its own software identifier, and sets its own test scenarios on its own calendar. Indiana approved my registration on 8 October and issued a software ID. Other states will not look at me until spring. A vendor can be fully approved in one state and entirely absent in the state next door, and both facts sit behind the same number on the brochure.
Tax type matters as much as geography. Ohio’s income tax developer window had closed for tax year 2026 by the time I reached it. In the same week, Ohio’s Centralized Municipal Net Profit Tax program sent me a schema package, dated 7 October. One state, two programs, opposite answers. If your clients file Ohio municipal net profit returns, “we support Ohio” tells you nothing you can act on.
Published guidance is not always current, either. Utah’s developer materials read, to me and I suspect to anyone, as though no registration is required. I asked, and Utah confirmed in writing that registration and testing are both required. Oklahoma answered the same question on the same day by sending me a mandatory letter of intent, Form D-106, which its published pages had not led me to expect. A firm reading those materials would reach the conclusion I reached, and it would be wrong.
The deadlines that decide next season are mostly invisible from the outside, and several have already passed. Minnesota requires a letter of intent by 30 October and states plainly that late submissions are not accepted. It cannot be sent by email at all; it goes through the state’s e-Services portal, and a first-time provider creates the account as part of submitting. Miss it and Minnesota is closed for the season, whatever else is true about the software. South Carolina’s letter of intent has fallen in early to mid August in recent years, and the state will not review forms until that letter is approved. By the time a firm is comparing products in November, some of these doors have already shut.
It is worth noticing how a state sequences its requirements, because that is often what decides who can enter at all. Minnesota’s own assurance testing opens on 2 November, after its 30 October letter deadline, so the letter is not gated on federal testing evidence. Iowa runs it the other way: the letter of intent and the IRS assurance testing documentation are both due on 2 November. For an established vendor that is routine paperwork. For a new entrant it is the difference between a state being genuinely reachable and being open only on paper. And some states simply decline. Arkansas told me no, in writing.
All of which suggests a different set of questions for your next vendor conversation.
Ask which states and which tax types the package supports for the coming filing season, with approval dates, rather than in general. Ask what happens to a return the software does not cover: paper, a second product, or a client left to file alone. Ask how far ahead the vendor is willing to commit, and notice whether the answer refers to state deadlines or only to the vendor’s own roadmap. Ask what they did last season when a state program closed early or a schema arrived late, because that answer is the best available guide to how they will behave when it happens again.
None of this requires technical knowledge. It requires treating state coverage as a set of dated approvals rather than a product feature, and then asking for the dates. Vendors who have them are usually glad to share them. Vendors who do not will change the subject, which is useful information in itself.
The firms I know who were caught out by this were not careless. They asked a reasonable question and got a reasonable-sounding answer. The question was just slightly the wrong one.
===
Tyler Briskin, EA, is an enrolled agent in private practice in Pennsylvania and the founder of Crosstie Tax, where he is building tax preparation software for small practices, covering workpapers through to the return, with a beta planned for Q1 2027.
Sign in to get access to this free resource, and all of our whitepapers and reports.
Download this content today!
Register Now Already registered? Click here to Log In
