Case · B2B SaaS
7 months
Cost per SQL down 44% after rebuilding search around intent
A support-software company buying clicks on competitor terms. We rebuilt search around intent tiers, piped CRM stages back into bidding and cut the SQL cost by 44%.
Illustrative
Demos booked / month
+3.1×
40
→
124
Cost per SQL
−44%
€410
→
€230
Qualified pipeline
+€1.9M
7-month total
→
attributed in CRM
Illustrative demo figures for a fictional client. Results vary with margin, market, offer and spend.
01
Challenge
Ticketloom sells help-desk software with auto-routing for support teams of 10 to 200 agents. It spent €36k a month on Google Ads and LinkedIn and booked 40 demos a month, at €410 per sales-qualified lead.
About a third of search spend went to competitor brand terms. Those clicks were expensive and mostly came from people comparing prices or looking for a rival’s login page. Google optimized to form fills, so it kept finding more of the wrong leads: students, agencies and teams of three.
Sales had stopped trusting marketing’s numbers. Only 25% of leads became SQLs, and fewer than half of those booked a demo. No CRM data flowed back to the ad platforms, so nobody could say which keyword produced revenue.
02
Strategy
In week one we recommended switching off the two competitor campaigns, which took a third of search spend and had produced three SQLs the previous quarter. Aisha then rebuilt search into four intent tiers: problem terms like “support ticket backlog”, solution terms, category terms and a small, capped brand campaign.
Daan connected HubSpot to Google Ads and LinkedIn with offline conversion imports for MQL, SQL and opportunity stages, each with a value. After six weeks of data, bidding moved from leads to SQL value.
We also changed the demo form. Team size and current help desk became required, and teams under ten agents went to a free trial instead of sales. Clicks fell slightly, from 8,780 to 8,489 a month, but leads rose from 351 to 509.
03
Creative approach
LinkedIn was the second lever. Ticketloom’s old ads showed dashboards and feature lists, and looked like every other help-desk ad in the feed.
We wrote for the support lead at 8:55 on a Monday instead. The winning 1:1 static carries the hook “Your support queue, triaged before standup.” with a simple ticket list and one line of proof: auto-routing for teams of 10 to 200 agents. The call to action offers a 20-minute demo, not a vague “learn more”.
We tested it against feature-led and review-led versions, aimed at support and CX job titles at companies with 50 to 1,000 employees. It produced SQLs more cheaply than any other LinkedIn ad and became the template for the next ten statics: one pain, one moment in the day, one number.
Funnel
Where
the
extra
revenue
came
from.
Bar length uses a log scale so every stage stays readable. Illustrative demo data; results vary.
Results
The
numbers,
month
by
month.
Over seven months, demos booked grew 3.1×, from 40 to 124 a month. Cost per SQL fell 44%, from €410 to €230. The CRM attributes +€1.9M in qualified pipeline to paid over the seven months.
Monthly spend rose from €36k to €41k. SQLs doubled, from 88 to 178 a month, and the share of SQLs that booked a demo rose from 45% to 70%, because the form now filters out poor fits before sales ever calls.
Sales and marketing now review the same dashboard every month, and bidding learns from pipeline instead of form fills.
Illustrative demo figures; results vary.
Creative
The ads that did the work.
The ads from this account, as they ran. Videos play muted in view; open the creative library to hear them.
Open the creative library
Ticketloom · ad
Result −44%
Priya Anand
Demo testimonial · fictional person
Next case
Northdose
MER from 2.2 to 3.6 with a subscription-first funnel
Get the same teardown for your account.
Five working days, a 90-day plan, and a straight answer on whether a retainer makes sense.