Smart bidding starves on a single metro immigration budget
Automated bidding needs a steady diet of conversions, and one practice in one metro rarely supplies enough of them.
Immigration keywords are expensive and the qualified volume is thin. Split a modest budget across six campaigns and every one of them learns from almost nothing, which is when you see wild swings in cost per click week to week.
Two fixes work. Consolidate: fewer campaigns, more signal in each. And count a conversion that happens often enough to teach the system something, such as a call that lasts past a threshold you set, rather than only a signed matter.
Start with tight keyword lists and manual control while you learn what the search terms actually look like. Move to automated bidding once the conversion count is steady, not on day one because the interface suggests it.
Resist the urge to add campaigns every time somebody has an idea. An account with three well fed campaigns beats one with nine that are all guessing.
Run the Front Range and the rest of Colorado as two different buys
Metro clicks and mountain clicks want different offers, and mixing them hides what each one really costs.
The metro buy covers Denver County and out through Aurora, Lakewood, Arvada, Westminster, Thornton, Centennial, Littleton and Englewood, down toward Castle Rock. That campaign can sell an office appointment, because the drive is plausible.
The second buy is everything else in the state. Fewer searches, usually cheaper, and often no immigration counsel nearby at all. Sell a video consultation in the ad copy itself, because the reader's first question is whether they have to come to you.
Keep the budgets separate. Cheap clicks from outside the metro will otherwise flatter your average and disguise what a Front Range case is actually costing.
Set location targeting to people present in the area rather than people interested in it, unless you have deliberately decided to take inquiries from abroad. Either choice is defensible. Drifting into it by default is not.
Detention and deadline searches need their own ads and a phone that answers
An urgency campaign is an operations commitment before it is a bidding decision.
Those searches come at night, on weekends, and usually from a relative rather than the person in trouble. The caller will try two or three firms in a row and stop at whoever picks up.
If you cannot answer outside business hours, do not run the campaign. Paying premium prices to send frightened people to voicemail is the fastest way to waste a month of budget and earn a bad review.
If you can, give it its own campaign, its own budget and its own copy stating hours and languages. Route those calls to a named human, not to a general queue.
Keep the copy factual. Nothing that implies a result, no urgency language that reads as a guarantee. How far ad copy can go on outcomes is governed by the advertising rules that apply to you, and it is worth confirming with your own counsel before the ads go live.
A negative keyword list built in January is stale by March
Immigration search wording moves with the news, so the waste changes shape every few weeks.
Read the search terms report weekly, not quarterly. The recurring drains are predictable: people looking for a government office, people wanting help filling in a form themselves, job seekers hunting sponsorship, students doing research, and citizenship test practice.
Then there is the wave that arrives whenever policy is in the headlines. Volume spikes, the wording changes overnight, and half of it is people looking for information rather than representation.
Add negatives at the level that actually holds, and keep a shared document so the list survives a change of account manager. Review it against the previous month rather than rebuilding it from a template.
One line to stay off: do not bid to intercept searches for free legal aid or community clinics. It converts poorly and it looks exactly like what it is.
Make the call the conversion and let the form be the backup
Consumer immigration inquiries close on the phone, and the account should be built to produce a ringing phone.
Use call assets and call-focused campaigns on mobile, with the number visible in the ad. A worried reader on a phone does not want to type a paragraph into a box.
Forms still matter for the employer side, where an office manager or a controller is browsing during the workday and wants to send a question rather than talk. Give that traffic its own ad group and its own landing page.
Set up call tracking so you know which campaign produced which conversation, and use call recording if you record at all. Recording and consent rules vary, and how they apply to your firm is worth confirming with your own counsel.
Count calls that clear a duration threshold you set. A thirty second call is usually a wrong number or a price question that never had a case behind it.
The ad platform never sees the signature, so keep the ledger yourself
Cost per lead is the number that makes bad accounts look healthy for six months.
Keep a monthly sheet: spend by campaign, calls, consultations booked, consultations attended, matters signed, and fees collected, with case type attached to each. It can live in a spreadsheet. It cannot live only in the ad account.
If your case management system supports offline conversion import, feed signed matters back so the bidding learns from outcomes rather than form fills. If it does not, the manual ledger still tells you where to move money.
Accept the lag. A campaign that looks expensive this month may be the one producing retainers next quarter, so judge it on a window long enough to include the signing, not on a calendar month.
Pace spend against consultation capacity. Buying more calls than the calendar can absorb produces a backlog, a slower callback and a lower close rate, which is a self inflicted wound the reports will blame on the ads.
Questions we actually get
- What should we budget for immigration ads in Denver?
- There is no honest number to quote you without seeing the auction for your case types, and any agency naming one in a first call is guessing. Work backward instead: what a signed matter is worth to you, how many consultations your calendar can absorb, and what share of clicks your intake can convert. That gives a floor worth testing, and the first sixty days are for measuring, not scaling.
- Should we run ads on Meta as well as Google?
- Immigration ads generally fall into a restricted category on the major social platforms, which limits how narrowly you can target and how ads can be worded. Social can still work for awareness and for reaching family members who are not searching, but expect broader targeting and a longer path to a call. Search should carry the account first.
- Are Local Services Ads worth it for an immigration firm?
- They can be, and they compete with your own search ads for the same click, so decide which one wins a given query rather than running both blind. The screening and verification process takes time and paperwork. Treat it as a separate channel with its own reporting, not as a line item inside search.
- Should we bid on other firms' names?
- It is technically possible and it is also a fast way to start a fight and pay a high price for a skeptical click. Defend your own firm name first, since competitors and directories will otherwise buy it out from under you. Anything involving another firm's name in ad copy is worth clearing with your own counsel against the advertising rules that apply to you.
- How soon will we know if paid search is working?
- You will see click and call data almost immediately, and it will tell you very little about revenue. The meaningful signal is signed matters, which arrive weeks after the click, so give the account a window long enough to include that lag. Anyone promising a lead volume or a cost per case before the account has run is selling a forecast, not a plan.
