Conversion Rate Optimisation for SaaS
Most SaaS conversion problems are not button colours. They are a demo form that asks fourteen questions, a pricing page that will not say a number, and a trial that takes four days to reach first value. We find where qualified traffic actually drops out, then rebuild that part of the journey around deal value rather than form fills.
What we can actually evidence.
How these are measured
- SaaS brands served:
- n = 30 SaaS brands. company inception to August 2026. Count of distinct SaaS companies under a signed advisory agreement. Excludes one-off audits, workshops and prospect calls.
- Years in-house experience:
- n = 1 founding team. 2015 to 2026. Cumulative years spent in salaried, in-house growth and marketing roles at SaaS companies before founding Surge45.
- AI assistants tracked:
- n = 6 assistants. current monitoring coverage. ChatGPT, Gemini, Claude, Perplexity, Microsoft Copilot and Grok, plus the AI Overviews and AI Mode surfaces inside Google Search. A coverage fact, not a performance claim.
- Growth pillars, one system:
- n = 4 pillars. current advisory model. Discover, Convert, Attribute and Grow. A structural fact about how the advisory is organised, not a performance claim.
The gaps quietly costing you pipeline.
Traffic is growing and demo requests are flat, so the funnel is losing people somewhere nobody has located
Sales says the leads are unqualified, marketing says there are plenty of them, and both are looking at different numbers
The pricing page will not state a price, so buyers who cannot get one leave and are never counted
Trial sign-ups are healthy but activation is not, and nobody has measured time to first value
Testing happens, but on low-traffic pages where no result could ever reach significance
The demo form asks for information sales does not use, and every field costs completions
How we drive compound growth.
Find the leak before touching anything
A full-funnel drop-off map by channel and stage. Most CRO programmes start testing before they know where the loss is.
Optimise for deal value
A change that raises form fills and lowers qualified pipeline is a loss. We measure the thing sales is judged on.
Test only where a test can conclude
Below a traffic threshold an A/B test cannot reach significance. On those pages we make reasoned changes and say so, rather than performing statistics.
Research, not opinion
Session recordings, exit surveys and sales-call listening. The reason people leave is usually knowable and usually not what the team assumed.
Why SaaS conversion is a different problem
Ecommerce CRO optimises a single session towards a purchase. A SaaS buying cycle runs for months across many people and several systems, and that changes almost everything about how the work should be done.
- The conversion event you can see, a form fill, is months away from the revenue event that matters
- Multiple people at the same account convert separately, which corrupts a naive per-visitor calculation
- The highest-value change is often removing a step rather than optimising it
- Self-serve and sales-led motions on the same site pull the page in opposite directions
- Activation and time to first value belong to conversion, but usually sit with product and go unmeasured
- Test velocity is capped by traffic, so most SaaS sites can run far fewer valid tests than their agency proposes
Where the work usually is
Form and qualification design
Every field costs completions. We audit which answers sales actually uses, and remove the rest or move them after the booking.
Pricing page honesty
A pricing page that refuses to indicate a number loses the buyers who cannot get budget without one, and they leave without a trace.
Trial activation and time to first value
Measure how long it takes a new account to reach the moment the product becomes useful, then remove the steps before it.
Journey design for self-serve and sales-led
Two motions on one site need two paths. Forcing both through the same page loses one of them.
Page speed and mobile experience
Unglamorous, measurable, and still the largest fixable loss on a meaningful share of SaaS sites.
Qualitative research
Exit surveys, recordings and listening to sales calls. This is where hypotheses worth testing come from.
How a CRO engagement runs
Instrument and baseline
Establish what can currently be measured, fix the gaps, and record the drop-off at every stage before changing anything.
Locate the largest loss
Quantify each leak in pipeline terms. The biggest percentage drop is often not the biggest revenue loss.
Research why
Recordings, surveys and sales-call listening on the specific step that is losing people, rather than site-wide.
Fix the obvious first
Removing an unnecessary form field does not need an A/B test. We separate the changes that need proving from the ones that need doing.
Test what genuinely needs testing
Sized against your traffic, with the required sample declared before the test starts and the result honoured either way.
Report in pipeline, then repeat
Every change reported in deal value, and the drop-off map refreshed so the next largest loss is the next piece of work.
Why SaaS teams choose us.
We locate before we test
Most CRO programmes start testing before establishing where the loss actually is.
Deal value, not form fills
The metric is qualified pipeline, agreed with sales before the work starts.
Honest about statistics
We will tell you when your traffic cannot support a valid test, rather than running one anyway.
Research-led hypotheses
Tests come from observed buyer behaviour, not from a list of best practices.
Joined to attribution
CRO reporting that reconciles with the CRM, so results survive a finance review.
We will say when it is not CRO
A qualified-traffic problem does not get fixed on the landing page, and we would rather redirect the budget.
Questions teams ask.
How much traffic do we need for this to work?
For a valid A/B programme, considerably more than most SaaS sites have on their highest-intent pages. Below that threshold the work is research-led: locate the loss, understand why, make reasoned changes and measure the before and after honestly rather than dressing it up as significance testing. We will tell you which category you are in at the start.
How long before we see results?
Instrumentation and the drop-off map take two to three weeks. Removing obvious friction can move numbers within a month. A test programme moves on your traffic volume, and a page that needs six weeks to reach significance needs six weeks. Anyone quoting a fixed timeline has not looked at your traffic.
Do you guarantee a conversion uplift?
No. Some tests lose, which is information rather than failure, and a programme that never reports a losing test is not reporting honestly. What we commit to is finding where the loss is, measuring it in pipeline terms, and reporting what each change actually did.
Should we put pricing on the site?
Usually at least an indication, yes. The buyers who need a number to start an internal budget conversation leave without contacting you, so the cost of hiding it is invisible in your analytics and real in your pipeline. The exception is genuinely bespoke enterprise pricing, and even then a starting band helps more than it hurts.
Does this overlap with our product team?
At activation, yes, and that is the point. Time to first value is a conversion metric that lives in the product, and treating it as somebody else's number is why it usually goes unmeasured. We work with product rather than around them.
More ways we help.
Lead generation
Demand and ABM strategy engineered around deal value, not raw form fills.
Learn morePaid & performance
Google, LinkedIn and programmatic spend directed for pipeline ROI.
Learn moreLifecycle & nurture
Email and lifecycle sequences that move leads from interested to revenue.
Learn moreMarketing analytics
The measurement layer that makes every conversion change arguable in pipeline terms.
Learn moreSaaS CAC optimisation
Lower acquisition costs through channel and conversion efficiency together.
Learn moreRetention
The pillar that decides what a converted customer is actually worth.
Learn more