Trades & Home Services
3.8x ROAS in 12 weeks
SaaS / B2B
-41% CPL, +18% MRR contribution
FieldRoute — -41% CPL via full-funnel rebuild
The challenge
FieldRoute had solid top-of-funnel numbers. Monthly unique visitors sat above 14,000, paid traffic was consistent, and the sales team had no shortage of inbound activity. The problem was the ratio: at $147 per qualified lead, with an average sales cycle of 45 days and a conversion rate from lead to paid customer of around 18%, the unit economics were under serious pressure. The growth targets weren't reachable at that CPL.
The marketing team's instinct was to look for a better-performing paid channel — move budget from Meta to LinkedIn, or try programmatic display. Before doing that, we asked to see the full picture.
A funnel audit across the eight months of available data told a different story. Click-through rates on Meta were actually fine — above-industry-average for the vertical in most ad sets. The conversion rate from ad click to landing page lead was where things broke down. Traffic was landing on a page that opened with a product feature list, a 9-field form, and a hero headline that described the tool rather than the problem it solves. Visitors were bouncing at 71% within 15 seconds.
Beyond the landing page, there was a second leak the team hadn't quantified: trial sign-ups who didn't convert to paid within the first seven days. FieldRoute had a 14-day free trial, and roughly 62% of trial users who didn't convert in the first week were simply going cold — no touchpoint from the business, no email sequence, no reason to come back. Sales were following up manually on high-value prospects, but the long tail of mid-market leads was being left entirely to chance.
Meta Ads had its own structural problems. Fourteen active ad sets were running, several overlapping on audience segments, with static image creative that hadn't been refreshed in four months. The algorithm was effectively competing against itself, and the creative fatigue was visible in the frequency and relevance score data.
---
What we did
We structured the engagement in three parallel workstreams, with a clear priority order: fix the landing page first, because improving conversion rate multiplies the value of every dollar already being spent on traffic. Then restructure paid. Then build the nurture layer to recover what the funnel was losing at the back end.
CRO sprint
We ran two landing page variants simultaneously using a simple A/B framework. The control was the existing page. Variant A changed the hero headline from a product descriptor to a problem-outcome framing. Variant B tested the same headline change plus a shorter form (4 fields instead of 9) and moved the customer logo row from the bottom of the page to immediately below the fold. We needed statistical significance across at least 200 conversions per variant before calling a winner, which took four weeks at the existing traffic volumes.
Variant B won clearly — conversion rate moved from 2.8% to 5.1%. The form length change had more impact than the headline, which was counterintuitive but consistent with what we see across B2B SaaS landing pages: friction reduction at the form stage is often the single biggest lever. The logo placement change lifted time-on-page by 22%, which suggests it was reducing bounce before the visitor reached the form.
Meta Ads restructure
Fourteen ad sets became four. We consolidated audience segments to eliminate overlap, let Advantage+ audiences handle broad prospecting (with strong exclusions: existing customers, current trial users, recent lead form completions), and rebuilt retargeting as its own campaign rather than mixing it with prospecting.
Creative was the bigger change. The existing static image creative was four months old with frequency above 3.2 across the warm audiences — visible creative fatigue. We tested video creative with a 6-second hook addressing the primary pain point directly, cut from existing product demo recordings rather than produced from scratch. The stronger-performing hook had a 34% higher thumb-stop rate and a 28% lower CPM in the first two weeks, which we interpret as the algorithm rewarding higher relevance.
Email nurture sequence
The six-email sequence for un-converted trial users was built from the ground up, with each email grounded in a specific objection drawn from two sources: a review of the sales team's CRM notes from deals lost in the prior quarter, and a short survey sent to a sample of trial users who'd churned without converting. The objections were predictable but specific: "I didn't have time to properly evaluate it", "I wasn't sure how it would work with my existing CRM", "I needed sign-off from someone else".
Emails one and two addressed time-to-value: a quick-start checklist and a 3-minute video walkthrough of the single feature with the clearest ROI for a field service operator. Emails three and four addressed the integration concern with a step-by-step CRM integration how-to. Email five was a case study from a comparable customer with a comparable problem. Email six was a direct commercial prompt: a limited-time offer to extend the trial by seven days for anyone who replied with a question.
The sequence ran to all trial users who hadn't upgraded by day 7. By the end of the first full month of the sequence running, 34% of previously cold leads had re-engaged in some form — clicked through, replied, or converted to paid.
Retargeting
A three-touch Meta retargeting sequence ran to everyone who'd hit the landing page without converting (90-day window). Touch one: a short video reframing the problem, no direct sell. Touch two: a customer proof post with actual results from a named customer, formatted as a simple card. Touch three: a direct response ad with the trial CTA. Sequence frequency was capped at 2 per week per person to avoid burning the audience.
Google RLSA bid uplifts of 40% were applied to visitors who had been to the site in the last 30 days and were searching high-intent terms — allowing us to prioritise budget on the warm audience without changing the campaign structure.
---
The results
Eight weeks from the audit to the first meaningful CPL improvement. Running controlled A/B tests takes time — but it's grounded in real data rather than a gut-feel redesign that might or might not work.
The headline result was a 41% reduction in cost per qualified lead, achieved without reducing ad spend. The budget was the same; the output changed. Landing page conversion rate went from 2.8% to 5.1%, which means the same traffic volume was generating nearly twice as many leads. CPL fell not because we found cheaper clicks, but because each click was more likely to convert.
MRR contribution from paid channels grew 18% over the three months following the restructure. That number reflects both the volume increase from better conversion rates and the revenue recovered from the email nurture sequence — 34% of cold trial leads who were previously being written off.
The Meta Ads restructure had an efficiency gain worth calling out separately: reducing from 14 ad sets to 4, with proper audience exclusions, also reduced the internal auction competition where FieldRoute's own ads were effectively bidding against each other. CPM on the core prospecting campaign dropped 19% in the six weeks after consolidation.
The lesson isn't complicated: the acquisition channels were performing within a normal range. The problem was everything that happened after the click. That's where the CPL improvements came from, and that's where they always come from in situations like this.
---
> "Every time we brought up our CPL numbers, the conversation would go back to 'you need more budget' or 'it's a competitive market'. What we actually needed was someone to look at what was happening after the click. That's where everything was falling apart." > > — Daniel H., Head of Growth, FieldRoute ---
Ready to find out what your account is actually doing?
We audit accounts every week. If yours has similar symptoms, we’ll find it.
