Split the account by who signs the check, not by visa category
An individual paying from savings, an employer paying from a budget and a relative paying from abroad are three different buyers, and one campaign hides all three.
Family-based work is usually bought by somebody spending their own money and comparing two or three firms. Employment work is bought by an HR generalist or an office manager who wants a process and an invoice. Humanitarian matters often reach you through a relative or an advocate who is not the client at all.
Each wants a different landing page, a different opening line and a different next step. The individual wants the fee. The employer wants to know your turnaround and whether you can handle several files at once. The relative wants to know whether you will even speak to them.
Structure should follow that split: separate campaigns, separate budgets, separate negative lists, separate reporting. Blended cost per lead across all three is a number with no decision attached to it.
Which one you scale is a capacity question. Buying employer work you cannot staff by Friday is an efficient way to make an enemy in Brentwood.
A circle drawn around The Gulch buys tourists and misses Mount Juliet
Target counties as separate units, because a radius from a downtown office covers the wrong people twice over.
Downtown carries a tourism and short-term rental economy, which parks a large number of people inside your radius who have no address in Tennessee and no interest in a consultation.
A radius from the same office also clips Mount Juliet, Hendersonville and Murfreesboro at the edges, which is where a great many recent arrivals settle and where nobody has a lawyer already.
Run Davidson, Williamson, Rutherford, Sumner and Wilson as distinct geographic targets with their own bid adjustments and their own reporting rows. The differences between them show up quickly and they are actionable.
Decide deliberately about people searching from outside the country on behalf of somebody here. Location of interest is a real setting with real volume. If you want that traffic, give it its own campaign and its own page rather than letting it distort a local one.
Free and consulate queries are the quietest way to lose a month of budget
Build the negative list before the keyword list, because immigration terms attract enormous volumes of people who are not buying.
Searches for free legal help, legal aid, pro bono clinics, consulate and embassy appointments, application status lookups and document translation or notary services all convert into staff time and no fee. They are cheap clicks, which is exactly why they accumulate.
Broad match on any immigration term will find you every adjacent conversation in the country. Exact and phrase match earn their place in this vertical, and the keyword list should be narrower than instinct suggests.
Review the search terms report weekly for the first two months, then monthly. Nobody enjoys the job. It remains the highest-value hour anyone spends in the account.
Add exclusions for matter types you do not take. A click for work you will refer out costs the same as a click for work you will sign.
Decide whether the consultation is paid before you write a single ad
The offer sets your allowable acquisition cost, your ad copy and your no-show rate, so settle it first.
A paid consultation filters. Fewer people book, a higher share attend, and everyone who arrives has already paid you once, which changes the conversation about a retainer.
A free consultation fills the calendar and moves the filtering onto your staff. It works when you have the capacity to absorb it and a disciplined way to end conversations that are not matters.
The failure mode is running paid traffic to a free consultation the firm quietly resents, then blaming the ads for lead quality. The traffic did what it was asked to do.
Put the price in the ad and on the landing page either way. Anyone who objects to a fee in the ad was going to object to it on the phone, and finding out before the click is cheaper.
An account you cannot listen to is an account you cannot fix
Most immigration inquiries arrive as calls, and unrecorded calls make paid performance unreadable.
Use tracking numbers at campaign level as a minimum. Dynamic number insertion is better, because it tells you which page was open when the call started.
Recording matters more than counting. A thirty second wrong-country call and a nine minute call that ends in a booked consultation both register as one conversion, and feeding both to the platform as equal will push the bidding toward the wrong people.
Score every call on a short scale: matter we take, matter we do not take, existing client, spam. Send only the first category back as the conversion the platform optimizes toward, and the account starts hunting the right traffic.
Call recording carries notice and confidentiality obligations. Worth confirming the setup with your own counsel before switching it on.
Judge the spend on attorney hours sold, not leads delivered
Lead count is a supply number, and an immigration practice is constrained by attorney and paralegal time.
A campaign producing work you cannot staff is a cost rather than a win. Pace the budget against the hours you can actually sell in a month, and hold some back for the weeks when demand moves for reasons outside anyone's control.
Track four figures per campaign: inquiries, consultations attended, matters signed, and fee value of those matters. Everything above that is diagnostic detail.
Report on cohorts by month of first contact rather than month of spend. Otherwise every review compares this month's invoice against signings driven by clicks bought two months ago, and the account gets cut for the wrong reason.
Nobody can quote you a cost per signed matter in advance for this practice area. What you can do is measure it honestly for ninety days and cut whatever loses the argument.
Questions we actually get
- How much should a Nashville immigration firm budget to start?
- Enough for one campaign, one payer type and one county group to accumulate a readable number of clicks inside a month. Spreading a small budget across family, employment and humanitarian work in five counties produces five unreadable rows. No agency can tell you the right figure before it sees your consultation fee and your average matter value, and any that quotes one blind is guessing.
- Should we run Meta alongside Google search?
- Search intent is where immigration budgets usually earn first, because someone typing a question is already in the problem. Paid social for this category sits under additional platform restrictions that change periodically, so current policy is worth checking before building anything. Social generally suits employer-side and community awareness work more than it suits urgent individual matters.
- Do we bid on other firms' names?
- It is technically possible in search and it is expensive, because the searcher already has somebody in mind. Attorney advertising rules and platform trademark policy both apply to how you word it, and both are worth confirming with your own counsel. Defending your own firm name is usually the better spend.
- How do we reach people searching from abroad?
- Run it as a separate campaign using location of interest rather than location of presence, and send it to a page written for someone who cannot walk into your office. Mixing that traffic into a local campaign will wreck your county reporting and your cost figures at the same time.
- How soon can we tell whether the account is working?
- You can read click and call quality inside two or three weeks. Signed matters take considerably longer, so build the reporting around cohorts by month of first contact and look at consultations attended as the early indicator. Deciding an account has failed before the first cohort matures is the most common expensive mistake in this category.
