Business Immigration Case Management Software: Who Is Actually Watching the Expiry Dates?

Why business immigration breaks the case tools built for it

It is a Tuesday in March. A paralegal has three things open: the myUSCIS organizational account for cap registrations, the DOL FLAG portal where a prevailing wage request has been sitting for months, and a spreadsheet with a tab per case type. An engineer's H-1B validity ends in 47 days. The spreadsheet says 78, because someone typed the date they expected rather than the one that arrived, and the approval came back truncated to the end of the labour condition application. Nobody caught it, because nothing in that spreadsheet derives a date from a document. A human typed it, and humans type what they assume.

That is the actual shape of the work. Employment based immigration is not a matter you open and close. It is a chain of dates hanging off a person: an LCA validity period, an H-1B validity period, a 60 day grace period after termination, a priority date waiting on the visa bulletin, an employment authorization document expiring while the adjustment application sits, a STEM OPT window, a PERM recruitment quiet period. Every one of those is derived from a PDF that arrived in a shared inbox. Miss one and a person loses the right to work, a project loses its lead engineer, and someone has a malpractice conversation with their carrier.

What INSZoom, Edge, Docketwise and Tracker actually leave you doing

None of these is bad software and we would not pretend otherwise. INSZoom and LawLogix Edge carry deep form libraries and questionnaire flows, and a mid sized practice doing conventional petition work will get filings out the door with either. Docketwise is the most pleasant of the group for a smaller firm and its form filling is genuinely fast. Tracker Corp is built for corporate teams and pairs immigration with I-9 administration.

They stop in the same four places. They are organised around a case rather than a person with a continuous status history, so answering whether this employee has ever had a gap means reading files rather than querying. Receipt notice handling ends with a human opening an I-797 PDF and typing a receipt number, which makes your dates of record exactly as reliable as that keystroke on a Friday afternoon. The corporate side gets a portal the firm controls, so a global mobility lead cannot join sponsorship data to headcount, wage bands and worksite records in Workday without exporting to Excel. And pricing runs per case or per sponsored employee, which quietly rewards tracking less. For a compliance function, that incentive is backwards.

Problem 1: every date that matters is derived, and yours is typed

An H-1B extension can be filed up to six months before the current validity ends, which means the task to start gathering evidence fires long before the date anyone has written down. The six year limit interacts with AC21: a pending PERM or labour certification filed 365 days or more before the sixth year buys one year increments, an approved I-140 with an unavailable priority date buys three year increments. A material change in worksite triggers an amended petition, which has been the position since Matter of Simeio Solutions. None of that is a date you type. All of it is a computation over the documents you already hold.

A custom build stores the source notice, extracts validity dates from it, and derives every downstream deadline from a rule per status type. Alerts fire at 180, 120, 90, 60 and 30 days against the derived date, not against a cell. When a notice arrives with a shorter validity than requested, the timeline shifts and the alerts move with it, which is the single behaviour a spreadsheet can never have.

Problem 2: one person's data, retyped into nine government forms

The I-129 with its H classification supplement, the ETA-9035 labour condition application, the ETA-9141 prevailing wage request, the ETA-9089, the I-140, the I-485, the I-765, the I-131, the G-28. The same legal name, the same employer identification number, the same worksite address, the same SOC code and wage level, entered again and again. USCIS retires form editions and rejects filings submitted on an expired one, so the risk is not only transcription, it is filing a perfectly correct petition on a stale PDF.

What a build does: one canonical person record and one employer record, versioned form templates with an explicit field map, and packet generation from that single source. The step operators actually value is the diff. Before filing, the system compares this petition's generated fields against the last approved filing for the same person and shows what changed. A job title that quietly drifted from Software Engineer to Engineering Manager is now a decision someone makes on purpose rather than a discrepancy a request for evidence finds for you.

Problem 3: the public access file is assembled after someone asks for it

The LCA rules require the public access file to be available within one working day of filing, holding the certified application, the wage rate, the actual wage memorandum, proof that notice was posted, and a summary of benefits. In practice, at most firms, it gets assembled during a Department of Labor investigation, from memory, three years later, for a worksite the company no longer occupies.

A build makes it a byproduct instead of a task. Certifying an LCA triggers file assembly, posting notices are generated with the correct worksite address and posting window and their dates recorded when marked complete, and the benefits summary is pulled from the employer record rather than a rediscovered PDF. When an investigator asks, you produce a file that was complete on day one because nobody had to remember to build it.

Problem 4: evidence chasing is the job, and nobody has automated it

Degree evaluations, transcripts, pay stubs, support letters, organisation charts, end client letters for third party placements, itineraries for multi site roles. The attorney hours in a petition are not legal analysis, they are following up with an employee in a different time zone about a credential evaluation. A build gives each petition type a checklist, opens two portals against it, one for the beneficiary and one for HR (Human Resources), reminds on a schedule, and refuses to route a file to an attorney until the checklist gates pass. Document extraction earns its place here on two jobs specifically: reading incoming I-797 notices to pull receipt number, notice type and validity dates, and reading a request for evidence to classify the issue raised and calculate the response deadline the day it lands rather than the day someone opens the envelope.

Problem 5: HR changes something and nobody tells the immigration team

A worksite move, a title change, a salary that drops below the LCA wage, a reduction from full time to part time, a termination. Each has an immigration consequence and each happens inside an HR system that has no idea. Workday, SuccessFactors or BambooHR events feed a review queue, and a termination starts the 60 day grace clock plus a petition withdrawal task automatically. For a corporate mobility team this integration is usually the entire reason to build rather than buy, because no law firm product will ever sit inside your HRIS event stream.

What this costs and how long it takes

Across the 2,000-plus projects Digital Heroes has delivered, this category prices predictably. A first release with the person level status timeline, the derived deadline engine, notice ingestion and the evidence portals runs $55,000 to $120,000 and ships in 12 to 16 weeks. A full platform adding generated form packets across the main petition set, public access file automation, HRIS integration, client and matter billing, and a corporate reporting view runs $160,000 to $400,000 phased across 7 to 12 months.

What pushes it up: the number of form types you actually file, because every form is its own mapping and validation exercise. Third party placement work, since itineraries and end client documentation multiply the evidence model. Multi entity employers with different identification numbers and worksites. And migration, which is the one people underestimate: pulling a decade of cases out of INSZoom or Edge into a clean person centric model is discovery work, not a script.

What keeps it down: starting with H-1B and L-1 only, one filing office, and your top two employer clients. That covers the bulk of the deadline risk and teaches the model everything it needs.

Build versus buy, and when buying is right

Buy if you are a firm under roughly 200 concurrent cases with a conventional petition mix. Docketwise or Edge will cost a fraction of a build and a custom system would be an expensive way to have the same forms. Buy if you are an employer sponsoring under 100 people and happy for your firm to hold the record.

Build when two or more of these are true. You carry over 300 concurrent cases and your deadline discipline depends on one person's diligence. You are an employer who needs immigration status joined to HR data for workforce planning, and exports are not a workforce plan. Your practice has a specific niche, cap exempt research employers, rural health placements, entertainment visas, where the packaged questionnaire flow fights you. You have been through a DOL or ICE inspection and improvised. Or you want to serve employers with a portal that carries your brand and your service model, which is a product decision, not a software decision.

How to choose a developer

Ask them to model status on a whiteboard before you sign. The right answer draws a person with a sequence of authorization periods, each with a basis, a source document and a derived expiry, plus dependent statuses hanging off it. A developer who draws cases with a due date field has built a task tracker and is about to learn immigration on your budget.

Ask how they handle a notice that grants less time than requested, because that single scenario tests whether dates are derived or stored. Ask what they have integrated: FLAG and myUSCIS have no friendly public API, so ask specifically how they plan to handle notice ingestion and case status checks, and be suspicious of anyone who promises deep portal automation without explaining the mechanics.

Ask about data ownership and get it in writing before kickoff. You should hold the repository, the cloud accounts and the unrestricted right to hire someone else. At Digital Heroes the client owns the code from the first commit. In a practice where the data is people's right to work, a vendor who holds your schema is holding more than software.