The SaaS Metrics PPC Agencies Often Get Wrong

Most SaaS PPC accounts don’t fail because the ads are bad.

They fail because the wrong numbers are being celebrated.

CTR looks healthy. CPL is dropping. The dashboard is reassuringly green. And yet… sales quality is inconsistent, CAC creeps up quarter by quarter, and the board starts asking awkward questions about whether paid media is actually doing anything useful.

This is rarely a bidding problem. It’s a measurement problem.

SaaS is structurally different from ecommerce and lead-gen-for-everyone-else. Longer sales cycles, multiple stakeholders, delayed revenue, expansion dynamics, all of that breaks simplistic PPC metrics if you treat them at face value.

Below are the SaaS metrics PPC agencies most often misinterpret, misuse, or lean on far too heavily and what to look at instead if you actually want efficient growth.

‍

First: why SaaS metrics are so easy to get wrong

SaaS sits in an awkward middle ground:

• Not transactional enough for instant ROAS.
• Not vague enough for “brand uplift” hand-waving.
• Dependent on post-click behaviour that PPC platforms don’t see by default.
  ‍

That means SaaS performance lives across:

• Ad platforms
• Analytics
• CRM
• Sales behaviour
• Retention and expansion data

Any agency that treats PPC reporting as “what happened in Google Ads” is only seeing a slice of reality. And that’s how well-meaning optimisation quietly drives the wrong outcomes.
‍

1. Cost per Lead (CPL): the most misleading metric in SaaS

CPL is attractive because it’s clean. One number. Easy to optimise. Easy to show progress.

It’s also the fastest way to optimise your way into a terrible pipeline.

Why?

Because all leads are not equal in SaaS, not by a long way.

A £40 lead from a junior researcher at a 10-person startup is not “better” than a £180 lead from a VP at a 500-person company. But CPL treats them as identical.

Where agencies go wrong:

• Optimising bids and targeting to lower CPL
• Broadening keywords and audiences to inflate volume
• Removing friction that was quietly qualifying intent

What to look at instead:

• Cost per qualified action (e.g. SQL, sales-accepted demo)
• Cost per opportunity created
• Pipeline value per £ spent

If your CPL halves but your demo-to-close rate collapses, you didn’t improve efficiency, you just bought cheaper noise.
‍

2. CTR: a proxy metric pretending to be a performance metric

Click-through rate tells you whether an ad is appealing, not whether it’s commercially useful.

In SaaS, high CTR often correlates with:

• Overly generic pain statements
• Curiosity-driven messaging
• Broad problem framing that attracts non-buyers

That’s fine before product–market fit, when learning matters. After PMF, CTR without context is a vanity metric.

Where agencies go wrong:

• Optimising copy purely for engagement
• Chasing “winning ads” based on CTR alone
• Assuming higher CTR = better targeting

What to look at instead:

• CTR by intent cluster (category vs use-case vs competitor)
• Post-click behaviour (bounce rate, time to first interaction)
• Conversion rate relative to CTR

A lower-CTR ad that attracts fewer but better-qualified clicks is often doing far more work for the business.
‍

3. Conversion rate (CVR): when “improvement” hides the real issue

Conversion rate feels sensible. If more people convert, the campaign must be better. Except in SaaS, CVR is highly sensitive to what you define as a conversion.

If your primary conversion is:

• “Book a demo”
• “Contact sales”
• “Start trial”

…then CVR is only meaningful if those actions represent roughly the same level of intent over time.

Where agencies go wrong:

• Changing forms, CTAs, or gating without recalibrating expectations
• Optimising to softer conversions to inflate CVR
• Comparing CVR across campaigns with very different intent profiles

What to look at instead:

• CVR by funnel stage
• Down-funnel conversion (demo → opportunity → close)
• Conversion quality signals passed back to platforms

A rising CVR paired with falling close rates is a red flag, not a win.
‍

4. ROAS: borrowed from ecommerce, awkward in SaaS

ROAS is brilliant when revenue happens at checkout.

In SaaS, it’s usually either:

• Incomplete
• Modelled
• Or based on first-touch assumptions that collapse under scrutiny

That doesn’t mean revenue metrics are useless, it means ROAS is rarely the right one.

Where agencies go wrong:

• Forcing ROAS models onto long sales cycles
• Using first-payment revenue as a proxy for customer value
• Ignoring retention and expansion entirely

What to look at instead:

• CAC payback period
• Contribution margin after acquisition
• LTV:CAC ratio by segment

These metrics reflect how SaaS businesses actually make money, over time, not at the moment of the click.
‍

5. “Blended CAC”: comforting, dangerous, and usually wrong

Blended CAC smooths everything into one number:

• Paid
• Organic
• Referral
• Expansion

It looks tidy. It also hides where paid media is helping, or hurting.

Where agencies go wrong:

• Using blended CAC to justify underperforming paid campaigns
• Losing visibility into paid-specific efficiency
• Avoiding hard conversations about channel-level performance

What to look at instead:

• Paid CAC vs blended CAC
• Paid CAC by ICP segment
• Paid CAC trend over time, not snapshots

Blended CAC is fine for board-level storytelling. It’s a terrible optimisation tool.
‍

6. Impression share: visibility without intent

Impression share sounds strategic. Who wouldn’t want to “own the category”?

In practice, it’s often a distraction, especially in SaaS categories with:

• Ambiguous search terms
• Research-heavy buyer journeys
• High competitor overlap

Where agencies go wrong:

• Chasing impression share on broad category terms
• Treating visibility as success
• Paying premiums for low-intent coverage

What to look at instead:

• Impression share on high-intent clusters
• Lost impression share due to rank on profitable terms
• Incremental pipeline generated per additional £ spent

Owning 80% of irrelevant searches is not market leadership. It’s an expensive theatre.
‍

7. “Number of leads”: the metric that refuses to die

This one refuses to go quietly, largely because it’s easy to understand.

But in SaaS, lead volume without context is meaningless.

Where agencies go wrong:

• Reporting lead count as a primary KPI
• Scaling spend to increase volume regardless of quality
• Ignoring sales feedback loops

What to look at instead:

• Leads that progress
• Leads by ICP tier
• Revenue per lead cohort

If sales don’t trust paid leads, the problem isn’t sales. It’s a measurement problem.
‍

8. Platform-reported conversions: incomplete by default

Google, LinkedIn, Meta, all report what they can see. None see your full sales cycle unless you make the effort.

Where agencies go wrong:

• Treating platform dashboards as truth
• Ignoring CRM reality
• Optimising based on incomplete attribution

What to look at instead:

• Offline conversion imports
• CRM-based reporting
• Channel influence, not just last touch

PPC platforms optimise to the signals you give them. Weak signals produce weak optimisation.
‍

The bigger issue: optimisation without context

Most SaaS PPC mistakes come from over-optimising local metrics without understanding the system they sit in.

Lower CPL.

Higher CTR.

More leads.

Better CVR.

All of those can move in the “right” direction while the business outcome moves the wrong way.

Efficient SaaS PPC requires:

• Commercial context
• Sales alignment
• Willingness to challenge “good-looking” numbers

And that’s uncomfortable, because it replaces tidy dashboards with nuanced conversations.
‍

A healthier SaaS PPC measurement stack

If you want PPC metrics that actually map to growth, anchor reporting around:

Primary

• Paid CAC (by segment)
• CAC payback
• Pipeline value from paid
• Win rate of paid-sourced deals

Secondary

• CVR by intent stage
• Cost per qualified opportunity
• Expansion behaviour of paid-acquired customers

Diagnostic

• CTR and CPL (as signals, not goals)
• Impression share (selectively)
• Platform-level efficiency trends

This doesn’t mean throwing away platform metrics. It means demoting them to supporting roles.
‍

Final thought: SaaS doesn’t need more dashboards, it needs better questions

Most PPC agencies don’t get SaaS metrics “wrong” because they’re careless. They get them wrong because they apply general PPC logic to a business model that behaves differently.

The right question is rarely:

“How do we lower CPL?”

It’s usually:

“Which paid activity creates customers who stick, expand, and justify the spend?”

Once you ask that, the metrics that matter become obvious, and the ones that don’t quietly lose their grip on decision-making.

For help building scalable PPC from an agency that nows SaaS inside and out, request a free proposal from Lever Digital today.

‍