Split the account by urgency before you split it by visa category
An emergency inquiry and a planned filing behave so differently that averaging them in one campaign makes both look mediocre.
Urgent work arrives with a deadline attached: someone detained, a notice with a date on it, a status about to lapse. Those searchers convert fast, tolerate a higher fee, and disappear within hours if nobody answers. Planned work, family petitions and naturalization among them, arrives calm, shops longer and often talks to three firms.
Run them as separate campaigns with separate budgets, separate ad copy and separate landing pages. Urgent campaigns get call extensions, tight daypart control and the highest bids you are willing to hold. Planned campaigns get lower bids, patient budgets and follow up built into the offer.
Only then split by category. Within each urgency tier, group by benefit type so the ad can match the phrase and the landing page can match the ad.
Blending them produces the classic bad report: acceptable cost per lead, terrible cost per retained matter, and no way to see which half caused it.
Bid the drive time, not the radius, because I-285 decides who shows up
A circle drawn on a map ignores the only thing your prospect is thinking about, which is how long the trip takes at the hour they can travel.
Radius targeting treats twenty miles north and twenty miles east as equivalent. Anyone who has crossed the Perimeter at five o'clock knows they are not. A prospect in Roswell and a prospect in Decatur may be the same distance from a Midtown office and an hour apart in practice.
Build geography in tiers instead. Tier one is the area that reaches your office comfortably outside peak hours. Tier two is reachable with effort and should be sold with a remote first offer. Tier three is everything else, bid low, sold entirely as a video consultation.
Write the ad to the tier. Someone outside the Perimeter should see a message about a video consultation and a same day call back, not an invitation to visit an office they will resent driving to.
Location bid adjustments then do real work. You are not fencing the caseload, which is federal anyway. You are pricing the friction that stands between a click and a person in a chair.
Most of your wasted spend is people looking for the government, not for a lawyer
Immigration keywords are crowded with navigational and do it yourself intent that will happily consume a legal budget.
Large blocks of search traffic in this category are people trying to reach a government website, check a case status, download a form, find an office address or look up a fee. None of them wants a consultation. Many of them will click a legal ad anyway because it sits at the top.
Build the negative list before the keyword list. Government agency names, status and tracking language, form download language, fee schedule language, office address language, job seeking language. Then keep adding weekly from the search terms report for the first two months, because that report is where the money actually goes.
Watch the two directions of employment intent. Someone looking for a job that sponsors and an employer looking to sponsor use overlapping words and are worth wildly different amounts. Separate them or exclude one.
Phrase and exact match do most of the protective work here. Broad match with a thin negative list is the fastest way to spend a legal budget on curiosity.
Cap the categories that eat consultation hours and pay the least
Paid search is a capacity purchase for a law firm, and some categories consume far more attorney time per dollar than others.
Every campaign you run competes for the same scarce resource: hours in front of a lawyer. A category that produces a lot of long consultations and few engagements is not a marketing success even when the cost per lead looks good.
Sit down with the calendar before setting budgets. Decide roughly how many consultations a week the firm can absorb and how they should be distributed across case types. Then set campaign budgets to buy that distribution, not to spend evenly.
Immigration demand also moves with news events that nobody schedules. Holding part of the monthly budget in reserve lets you respond when a category spikes instead of watching a daily cap throttle you on the day it mattered.
Pausing a campaign is a legitimate move. Firms treat it as failure. When intake is buried or an attorney is out, the spend should follow.
Sponsorship budgets get approved in Alpharetta office parks, not in the search auction
Employer side immigration work rarely starts with a search, so a search only account will keep telling you the segment does not exist.
A company deciding to sponsor is running an internal process. HR raises it, finance asks what it costs, an executive signs. By the time anyone searches, the firm has often already been chosen by a referral or a name someone recognized.
Reaching that buyer generally means paid social with job title and company size targeting, plus a real asset worth requesting: a process outline, a cost breakdown, a calendar of what happens when. Not a consultation booking form. They are not ready to book anything.
The geography for that buy looks nothing like the consumer one. Employers cluster along the Perimeter and up through Sandy Springs, Alpharetta and Johns Creek, and the person you want is at a desk during business hours.
Platform rules for legal and sensitive categories change, and some targeting options are restricted. Worth confirming current policy on each platform before building the buy, rather than after a disapproval.
Feed signed retainers back into the account or it will optimize for tire kickers
Automated bidding gets better at producing whatever you tell it to count, so counting form fills teaches it to find people who fill in forms.
Set the conversion actions in layers: inquiry, consultation booked, consultation attended, matter opened. Then import the later ones back into the ad platform so bidding learns from the outcome rather than the button click.
The lag is real. An immigration matter may take weeks between the click and the engagement, so the account is always optimizing on incomplete information. Build the reporting to wait, and resist judging a campaign on its first fortnight.
Report cost per attended consultation alongside cost per opened matter, split by urgency tier and category. Those two numbers will usually reorder your budget within a quarter.
Cost per lead should not appear in the monthly conversation at all. It is the number that makes a bad account look fine, and it is the reason so many firms conclude that paid search does not work for immigration.
Questions we actually get
- What should we expect to pay per click for immigration keywords in Atlanta?
- No honest answer exists without seeing your account and the categories you want. Legal clicks are among the most expensive in search, removal defense and employment based terms tend to sit higher than general terms, and prices move constantly. Any agency quoting a figure before an audit is guessing.
- Should we bid on our own firm name?
- Generally yes, if competitors or directories are showing above you on it. The clicks are cheap and the traffic is people who already decided to call you. Check first: if nobody is bidding against your name and you already hold the top organic result, the spend may be unnecessary.
- Is Google Local Services worth adding alongside search ads?
- For many firms it is worth testing, because it charges per lead rather than per click and appears above traditional ads. Verification requirements apply and availability varies by category and market, so confirm eligibility and current terms before planning a budget around it.
- How much budget does a paid search program need to produce usable data?
- Enough to buy a meaningful number of consultations per month in each campaign you are trying to evaluate. Splitting a small budget across six categories produces six inconclusive campaigns. Start with the one or two categories that matter most to the firm and widen once they are stable.
- Can advertising promise a case outcome?
- No. Ad copy and landing pages for a law firm carry professional responsibility rules, including restrictions on claims and required disclaimers. Worth confirming the specifics against State Bar of Georgia guidance or with your own counsel before anything goes live.
