Select Page
Case Study: Hero

Case studies / Legal

Criminal defence · Google Ads

Almost half of them wouldn't pick up the phone.

In a sample week, Chanco Schiffer Law received 12 calls and 10 form submissions: 22 leads on $611 of media, at under $28 each. The form half only exists because we gave people somewhere to go that wasn't a phone call. In criminal defence, a lot of people can't make one.

Client Chanco Schiffer Law
Practice Criminal defence
Engagement 1 year · $2,000/mo
Inherited from A previous agency
Case Study: Results
22Leads in one weekOn $611 of media
$27.77Cost per leadCalls and forms
45%Arrived by form, not phone10 of 22
3Ways to make contactCall, form, chat
Case Study: About the Client
Chanco Schiffer Law website
About the client

A criminal defence firm inheriting an account built to miss half its demand.

Chanco Schiffer Law practises criminal defence, for clients who are frequently in circumstances where a phone call is the last thing they can make.

We were brought in after Google Ads had been run by a previous agency, on automated bidding pointed at the firm's general website with a single route to get in touch.

Case Study: The Idea
The idea that shaped the account

A phone number is a filter. Sometimes an expensive one.

Criminal defence advertising is usually built around the call, and for good reason: urgency favours the phone. But someone facing a charge is often somewhere they can't talk about it. At work. In a house where family don't know yet. Sitting next to someone. An account that only accepts phone calls doesn't reduce that demand, it just never sees it.

Phone only

You see the people who can talk.

Call focused ads and a number on every page. It captures the urgent, the private, and the confident, and loses everyone whose circumstances make a phone call impossible right now, without ever showing you that they existed.

  • Captured Callers
  • Missed Everyone else
  • Visible in reporting No
  • Apparent demand Understated
Call, form and chat

You see the rest of them.

Three routes to the same conversation, so someone can reach a lawyer from a desk, a waiting room, or a bedroom without speaking out loud. In the sample week, nearly half of all leads came through a route that wasn't the phone.

  • Captured Callers and writers
  • Form share 45%
  • Cost per lead $27.77
  • Apparent demand Close to real
Case Study: Starting Position
Starting position

An inherited account, automated into the ground.

Chanco Schiffer Law came to us after Google Ads had been run by another agency without producing the criminal defence enquiries the firm needed.

The configuration explained most of it.

  • Automated bidding with nothing to learn from

    The account used an automated bid strategy. Those work by optimising toward conversion data, and in an account that isn't generating conversions reliably, there's nothing for the algorithm to learn from. It spends confidently in whatever direction it guesses.

  • Traffic pointed at the main website

    Ads sent people to the firm's general website rather than to pages built for the search that brought them. The visitor arrives with a specific charge and lands on a homepage.

  • One way to get in touch

    Everything funnelled toward a phone call, with no alternative for someone who couldn't make one at that moment.

  • $2,000 a month, in criminal defence

    A modest budget in a category with expensive clicks. Automated bidding and a generic destination are affordable mistakes at $20,000 a month. At $2,000 they are the whole budget.

Case Study: The Build
The build

Four changes, in this order.

The first two had to happen before automation could ever work.

01

Take back manual control

The automated strategy was removed and the account restructured around campaigns we could steer directly: search, call only, and local. Automated bidding is powerful once an account produces steady conversion data. Before that it's an expensive way to guess, and a small budget can't fund the guessing.

Spend directed rather than delegated.

02

Stop sending traffic to the website

Dedicated landing pages built for the searches driving the clicks, replacing the general site as the destination. A homepage asks a visitor to find their own answer; a landing page gives it to them in the first screen.

Ad and destination finally describing the same thing.

03

A/B test the pages properly

Several landing page variants tested against each other rather than one built on assumption. Layout, messaging, and where the contact options sat were all decided by what performed, over a year of running.

Page decisions made on evidence.

04

Offer three ways to make contact

Phone, form, and live chat, given equal prominence. Call tracking on every inbound so both routes were measurable. This is the change that revealed how much demand the phone only setup had been missing.

45% of leads arriving through something other than a call.

Case Study: A Week From The Account
A week from the account

$611, 22 leads, two routes.

  • Media spend $611
  • Tracked phone calls 12
  • Form submissions 10
  • Total leads 22
  • Cost per lead $27.77

That's a single reported week rather than a twelve month average, and it ran slightly above the account's usual weekly rate. It's included because the split inside it is the point.

Ten of those twenty two people had a legal problem serious enough to contact a defence lawyer and still chose to type rather than talk. On a phone only account, all ten would have been invisible. Not lost, exactly. Never counted.

It's worth comparing against our DUI defence account in Arizona, where 91% of leads arrived by phone and forms went to zero in the best month. Same broad category, opposite split, because that account was built around the call and this one wasn't.

Neither number is wrong. They're measuring what each account was set up to receive. Which is the uncomfortable part: your contact options don't just capture demand, they define what you can see of it.

Case Study: What Transfers
What transfers

Four things worth taking from this.

Automated bidding needs data first

Smart bidding optimises toward conversions. An account not producing them reliably gives the algorithm nothing to learn from, and it will spend the budget anyway.

Your homepage is not a landing page

Sending paid traffic to a general site asks the visitor to do the work of finding their own answer. Most won't, and the click is already paid for.

Count what a phone number excludes

In sensitive categories, legal, medical, debt, addiction, a meaningful share of people can't speak freely. Give them a way to write instead and see what appears.

Small budgets need manual control

Automation is a good tool with enough volume behind it. Below that, it's an expensive way to explore, and the exploration is your whole month.

Case Study: Services Used
Services used

What went into it.

Five of our services, over one year.

Case Study: More Work
Case Study: Final CTA
Next step

Is your account on automated bidding with nothing to learn from?

It's the most common thing we find in inherited accounts under a few thousand a month, and it looks like normal management until you check what the algorithm is optimising toward. Give us read only access and we'll tell you. Findings are yours to keep either way.

Read only access · No obligation · You keep your accounts