Local Services Ads or regular Search for a law firm: which one first?

Take a small family law firm with about $2,000 a month. LSA charges per lead instead of per click, and the Google Guaranteed badge does a lot of the selling, so the cost per lead can look great on paper. The catch is that some of those leads are just someone calling three firms to compare prices, and you have less say over which searches trigger your ad. Regular Search is more work because you pick the keywords, write the copy and build the landing pages, but you control intent and can actually measure it. I'd start with Search if you can track calls and form fills reliably, and only add LSA once you know what a qualified lead is worth to you. How are others handling the dispute process when LSA sends junk leads?

Agreed on the order. What I'd add is about the baseline. LSA's cost per lead only looks cheap if you know what a paying client is worth to you, and a Search campaign is the easiest way to find that out, since you can see which terms brought in cases that actually signed. Without that number you're comparing a per-lead price against a guess.

Disputes are another place where the feedback loop is weaker than people expect. In Search you can pull the search terms report and see exactly which queries burned money. In LSA you never see the queries, so the lead's own description is your only signal. Keep a simple log with the date, the lead, why you disputed it (wrong practice area, outside your service area, spam, duplicate) and whether it got credited. If the same reason keeps showing up, you have a pattern to put in front of Google, and that's a much stronger case than a general complaint.

Agreed on the log. The lead's own words are usually more useful than the category label, though. LSA doesn't show you a query, but the lead description and call notes tend to say what the person was actually after, so tag each lead with that phrase. After a few weeks you can see whether the junk is coming from one practice area that's leaking in.

Tagging by the lead's own phrase makes sense, but only if intake is consistent. How useful it is depends on volume. At a handful of leads a month you can read every one yourself and the pattern shows up within a week. At a hundred a month you need a rough tag set, or the notes just turn into noise.

A week is too short, honestly. At a handful of leads a month you're getting one or two a week, so after seven days you've read a couple of people and you're mostly seeing their individual situations. At that volume a real pattern usually takes six to eight weeks to show up.

I'd also tag the outcome next to each phrase. Two "divorce" leads that signed tell you more about where to spend than five that sounded promising and went nowhere, and at low volume those outcomes are the only signal worth acting on.

Six to eight weeks sounds right to me, and tagging outcomes is the way to do it. One rule of thumb I'd add: don't move budget off a phrase until it has produced at least three leads that reached a consultation, since one or two signed cases can easily be luck. At one or two leads a week, that's usually your real timeline anyway.

Three consults is a sensible floor, but family law makes the timing messy. Some people sign on the first call, and others sit on it for a month or two while they decide or gather paperwork. That means a six-week read can miss the phrases that convert slowly. I'd score each phrase on consultations booked in the early weeks, then recheck signings once those cases have had time to close.

I'd make the rule explicit so nobody reads a slow phrase as a dead one. Pick a review date, maybe 90 days out, or whatever your typical time from consult to signing is. Until then, judge phrases only on consultations booked, since that's the signal you can see early. Signings should only decide where the budget goes on that date. Without a fixed date, you end up either funding phrases that never convert or cutting ones that were just slow to close.