Imagine pouring thousands of dollars into acquiring new users, only to watch them quietly abandon your platform before ever converting. This is the reality for the majority of SaaS companies today, and the numbers are staggering. Research shows that 68% of businesses are making critical conversion decisions based on incomplete data, gut instincts, or outdated benchmarks.

The role of a conversion optimizer is no longer optional in a competitive SaaS landscape. It is a strategic necessity that separates high-growth companies from those stuck in a frustrating cycle of traffic without revenue. Yet most teams either misunderstand what optimization truly involves or lack the analytical framework to execute it effectively.

In this analysis, we will break down exactly why so many SaaS companies are optimizing blindly, what the underlying data reveals about common failure points, and how a structured approach to conversion optimization can dramatically shift your outcomes. Whether you manage a freemium product or a high-touch enterprise solution, the insights here will give you a clearer, more actionable path forward.

The Conversion Performance Gap Is Costing You More Than You Think

If your SaaS website is converting visitors at the industry average of 1.5%, you are not running a traffic problem. You are running a conversion problem, and the financial consequences compound every single month. 2026 benchmark data from PixelsWithin reveals that elite B2B SaaS companies convert visitors to leads at 8 to 15 percent, creating a delta that is not marginal. It is a 5x to 10x difference in pipeline generation from identical traffic volumes. For a company receiving 10,000 monthly visitors with a $12,000 ACV, the gap between 1.5% and 8% conversion represents the difference between 18 and 96 qualified leads per month, a spread that compounds into millions in unrealized ARR annually.

The mechanism behind this gap is process, not spend. Top-performing SaaS companies are 5x more effective at turning visitors into pipeline than their average counterparts, and the separating factor is systematic funnel optimization. These organizations track pipeline value, SQL quality, CAC by channel, and LTV:CAC ratios monthly. They treat funnel data as an operational input, not a reporting artifact. The companies sitting at 1.5% typically lack this infrastructure entirely; according to B2B lead generation research from Callbox, 79% of leads never convert to sales without proper nurturing, and 67% of lost sales opportunities stem directly from poor lead qualification, both symptoms of undisciplined funnel management.

Channel allocation makes the gap worse for most teams. The cross-industry average conversion rate sits at 6.6%, but SaaS-specific benchmarks expose significant channel variation. Email drives the highest conversion at 19.3%, while paid search delivers just 1.2 to 1.5%. Most SaaS companies are over-investing in paid acquisition without the conversion infrastructure to justify that spend, and paid search CPCs rose another 11% year-over-year in Q2 2026, making every unconverted click progressively more expensive.

At the scale of a $195 billion global SaaS market, even a one-percentage-point improvement in conversion rates represents eight-figure revenue opportunities for mid-market players. The companies capturing that upside are doing so through systematic funnel instrumentation, not by outspending competitors on acquisition.

Why Most Optimization Efforts Fail Before They Start

According to the Content Marketing Institute, 68% of B2B SaaS companies have no documented funnel optimization strategy. That statistic deserves a moment of consideration. The majority of teams running A/B tests, refreshing landing pages, and tweaking CTAs are doing so without a governing framework to evaluate whether any of it is working. They are accumulating activity, not building a system. Without documentation, there is no baseline, no hypothesis structure, no iteration cadence, and no way to distinguish a winning change from a random fluctuation. Disconnected tactics cannot compound into a competitive advantage; only a repeatable, measurable system can.

Attribution blindness is one of the most costly and least visible problems in conversion optimization. When teams cannot accurately connect conversions to the channels that drove them, budget reallocates toward whatever looks most active rather than what is actually performing. Consider a scenario where organic search is quietly driving 40% of qualified pipeline while a paid campaign generates high click volume but shallow intent. Without proper attribution infrastructure, the paid campaign appears more productive. Teams cut or starve the organic investment and scale the inefficient channel, compounding the performance gap month over month. Research indicates that teams with proper attribution infrastructure can reduce customer acquisition cost by 20 to 40%, not because they changed their tactics, but because they finally understood which tactics were working.

Funnel visibility is not a reward for doing optimization well; it is the entry requirement for doing it at all. The foundational step in SaaS conversion optimization is knowing exactly where visitors exit each stage of the funnel. Without that visibility, every test is directionally blind. You may be optimizing the wrong page, solving the wrong friction point, or measuring the wrong outcome entirely. A CTA change on a pricing page means nothing if the real drop-off is happening at the feature comparison stage three steps earlier.

This leads directly to the distinction most teams miss: the difference between activity and optimization. Running tests is activity. Systematically diagnosing conversion leaks through data analysis, user research, and controlled experimentation is optimization. The first creates motion; the second creates measurable outcomes. McKinsey data reinforces this: SaaS organizations that systematically optimize their funnels achieve 30 to 50% improvement in conversion rates. The operative word is "systematically," and it separates the 32% of companies building genuine optimization engines from the 68% still mistaking motion for progress.

The Four Conversion Optimizer Levers That Actually Move the Number

Most SaaS teams approach optimization the same way they approach a leaking pipe: they patch the most visible drip without checking whether the pressure loss is actually somewhere else in the system entirely. The four levers below represent a more diagnostic, infrastructure-first approach to conversion optimization, one that addresses the full funnel rather than isolated moments within it.

Lever 1: Funnel Stage Diagnosis

Before changing a single headline or CTA button, map precisely where your funnel is hemorrhaging. The instinct for most teams is to focus on top-of-funnel traffic volume, but the data consistently points elsewhere. Research shows that SaaS trial-to-paid conversion sits at a median of just 14.7%, meaning 85 of every 100 signups never generate a dollar of revenue. More revealing still, 61% of SaaS companies rely on a single welcome email and manual follow-up to manage that critical conversion stage, leaving the highest-value moment in their entire funnel effectively unmanaged. Conversion improvements are also multiplicative rather than additive: a 20% lift at acquisition combined with a 10% lift at the sales stage produces 32% total growth, not 30%. That compounding math makes accurate stage-level diagnosis disproportionately valuable compared to optimizing any single touchpoint in isolation.

Lever 2: Attribution-Informed Channel Allocation

Knowing where visitors drop off is only half the picture. Understanding which channels deliver buyers worth converting is equally critical. Organic search represents the dominant entry point for the majority of SaaS buyer journeys, yet last-touch attribution models systematically undercount its pipeline contribution by crediting conversion to whichever touchpoint happened last. The practical consequence is chronic underinvestment in SEO, precisely the channel responsible for most of your pipeline. The cost differential is substantial: organic trial acquisition averages $47 per signup versus $142 via paid channels. When 85% of trials fail to convert regardless of source, blended CAC waste can exceed $280,000 annually for a mid-market company without multi-touch attribution surfacing those inefficiencies. Moving to a linear or data-driven attribution model does not change your traffic; it changes which investments you make more of.

Lever 3: On-Page and Flow Optimization

The 2026 shift toward "show, don't tell" is measurable, not just directional. Case evidence from conversion-focused redesigns shows demo signup rates tripling and conversion rates jumping 300% when product value is communicated visually and within the first few seconds of a page visit. Free trials requiring a credit card convert at 30%, more than five times the rate of no-card trials, according to ChartMogul's January 2026 study of 200 B2B software products. Mobile traffic constitutes the overwhelming majority of landing page visits, yet desktop sessions convert meaningfully better. That gap is not a design problem; it is a flow problem, and most SaaS teams are not building mobile-specific sign-up experiences that remove the friction points unique to smaller screens and touch-based navigation.

The majority of conversion optimization frameworks treat a signup as a finish line. For SaaS, it is closer to the starting gun. Per Pendo's 2025 Product Engagement Report, only 34% of trial users ever reach an activation milestone. Of those who do, 48% convert to paid. Of those who do not, only 4.1% convert. The activation event is not a product metric; it is your most predictive conversion metric. Reaching the product's "aha moment" within 72 hours is the single strongest predictor of paid conversion, yet only 29% of trial users complete the activation sequence within that window, largely because onboarding flows are triggered by time rather than behavior. Companies deploying behavioral, automated trial nurture sequences improve trial-to-paid conversion by 20 to 28% and activate 3.1 times more users to that critical milestone. Improving trial conversion by just five percentage points, from 15% to 20%, translates to a 33% increase in ARR without adding a single new acquisition dollar.

The Infrastructure Layer That Connects All Four

Each lever generates signal. The compounding value comes from reading all four signals simultaneously. A funnel management dashboard that surfaces stage-level drop-off, channel attribution, on-page performance, and post-signup activation data in a single operational view transforms optimization from a quarterly exercise into a continuous discipline. Without that unified infrastructure, teams optimize one lever while remaining blind to regression in the others, and the multiplicative math works against them rather than for them.

Funnel Stage Diagnosis: Finding the Leak Before Fixing It

A structured funnel audit treats each stage as an independent diagnostic unit. The path from visitor to lead, lead to trial, trial to onboarding completion, and onboarding to paid conversion each carries its own failure modes and requires its own intervention logic. A team struggling with trial-to-onboarding drop-off needs an activation strategy, not a landing page rewrite. Mapping discrete conversion rates at each transition makes these distinctions visible and prevents resources from flowing toward the wrong fix. As Lucky Orange's funnel analysis framework notes, a single aggregate site conversion rate is fundamentally deceptive; the critical leak is almost always invisible in the top-line number.

Once stage-level rates are mapped, cohort-based analysis determines whether a problem is chronic or episodic. A chronic problem affects all users regardless of acquisition source or timing, pointing toward a structural product or messaging issue. An episodic problem clusters around a specific campaign, traffic source, or product release, meaning it has a definable cause and a definable fix. Without this segmentation layer, teams routinely treat episodic symptoms as systemic failures and redesign entire funnel stages in response to what was, in reality, a single underperforming campaign.

Real-time funnel dashboards replace the quarterly spreadsheet audit with a live diagnostic capability. Rather than discovering a conversion leak weeks after it begins compounding, SaaS teams can observe stage-by-stage drop-off as it develops and intervene before the revenue impact accumulates. According to Funnel.io's research on common funnel mistakes, data fragmentation is frequently the root cause of funnel misdiagnosis, which underscores why centralized, real-time visibility matters more than periodic manual review.

Benchmarking provides the calibration layer that transforms diagnosis into prioritization. Elite B2B SaaS teams achieve visitor-to-lead conversion rates of 8 to 15%, compared to the industry average of just 1.5%. That gap is not an abstract benchmark; it represents a concrete, measurable target that tells teams whether their current performance reflects a real optimization opportunity or simply reflects the upper bound of what their funnel model can deliver. Without stage-level benchmarks, optimization has no reference point, and improvement remains undefined.

Attribution Is a Conversion Optimizer Tool, Not Just a Reporting Function

Most SaaS teams treat attribution as a finance function: something you review after the quarter closes to explain where revenue came from. That framing is costing them conversions they never realize they lost.

Last-touch attribution assigns 100% of conversion credit to the final channel a buyer interacted with before converting, collapsing the entire upstream journey into a single event. The structural damage this creates is well-documented. Research from multi-touch attribution analysis indicates that last-click models misattribute between 30% and 60% of revenue, with B2B SaaS sitting at the high end of that range. Because 67% of SaaS buyers begin their journey through organic search, content and SEO channels consistently absorb this misattribution penalty. They generate the awareness that initiates the purchase journey, but they rarely hold the last click, so they appear to underperform on every dashboard that uses last-touch logic. Retargeting and branded search, which habitually intercept buyers already moving toward conversion, absorb inflated credit they did not earn. The budget consequence is predictable: teams cut demand-creation spend and double down on retargeting that only captures visitors already intent on converting, gradually strangling the top-of-funnel pipeline that makes retargeting viable in the first place.

The correction is not incremental. Moving to multi-touch, revenue-weighted attribution is one of the highest-leverage operational changes a SaaS growth team can make. In revenue-weighted models, closed-won deal value flows backward through the touchpoint path, assigning proportional credit based on each channel's measurable contribution to pipeline. McKinsey data shows that organizations implementing multi-touch attribution reallocate 18% to 22% of budget across channels and reduce customer acquisition costs by 12% to 19% as a direct result of more accurate spend decisions. Those are conversion efficiency gains, not reporting improvements.

Despite the evidence, only 24% of B2B organizations currently use multi-touch attribution, according to Gartner's 2025 survey data. The majority are still operating on models that systematically suppress their own conversion performance by misallocating budget away from the channels doing the heaviest lifting. In 2026, attribution sophistication is increasingly framed as growth infrastructure, not a reporting convenience, with leading teams connecting attribution outputs directly to pipeline stage data rather than surface-level conversion metrics.

FunnelKeeper's attribution intelligence layer operationalizes this shift. By connecting marketing touchpoints to pipeline outcomes at the channel level, it gives growth teams the clarity to allocate spend toward what actually drives conversion progression rather than what is easiest to measure in a last-touch dashboard. For SaaS teams serious about closing the conversion performance gap, accurate attribution is not the final step in optimization. It is the foundation everything else is built on.

2026 Trends Redefining What Conversion Optimization Means

The definition of conversion optimization is being rewritten in real time, and the 2026 shift is more structural than cyclical. Five distinct forces are converging to move CRO from a page-level discipline into a visitor-level intelligence function, and SaaS teams that miss this transition will find their optimization efforts producing diminishing returns regardless of how much testing they run.

AI-driven hyper-personalization has become the dominant CRO investment category. Tools are now dynamically adapting landing pages, CTAs, and pricing displays to individual visitor profiles using firmographic data, behavioral signals, and intent scoring. This is not A/B testing at scale; it is a fundamentally different operating model. According to 2026 CRO Year in Review research, programs implementing AI personalization saw an average 28% conversion lift alongside a 34% increase in average order value. By the end of 2026, more than 60% of CRO programs are projected to use AI for hypothesis generation, test analysis, or personalization, with manual-only programs falling measurably behind. The underlying driver is contextual and behavioral personalization rather than identity-based targeting, a direct response to third-party cookie deprecation that forces teams to build visitor intelligence from first-party signals.

Pricing pages have become an active conversion variable rather than a static information display. Personalized pricing adapted to company size, detected use case, or intent signals removes a specific and underappreciated friction point: the moment a visitor cannot locate themselves within a pricing structure and exits without converting. Static tiered models were designed for clarity, but they produce confusion when the visitor's context does not map neatly to the available tiers.

The "show, don't tell" principle is accelerating as the primary trial-conversion lever. SaaS brands embedding product walkthroughs, explainer clips, and testimonial videos directly into landing pages and trial onboarding flows are attacking time-to-value friction, which is the primary structural reason free trials fail to convert to paid subscriptions. Feature lists create comprehension work; demonstrations create confidence.

Behavioral and emotional signals are entering the analytics stack alongside traditional quantitative metrics. Scroll depth, session hesitation patterns, and engagement intensity are now being used to identify where conversion intent fractures rather than just where visitors exit. Per CRO software analysis for 2026, tools are automating pattern analysis from heatmaps and session replays to generate data-backed optimization hypotheses without requiring manual interpretation.

Attribution model sophistication is rising as a direct response to wasted spend. The shift from last-touch to multi-touch and revenue-weighted attribution reflects a market-wide recognition that conversion optimization produces misleading signals when it cannot accurately identify what caused the conversion. Full-funnel optimization programs are generating 3 to 5 times the revenue lift of page-level programs, and that gap is largely explained by measurement accuracy. Teams optimizing conversion rates without first resolving their attribution model are, in effect, tuning an instrument that is not reading the correct notes.

Vibe-Coded Apps and the Unique Conversion Optimization Challenge

Vibe-coded apps represent a genuinely new category in the SaaS landscape, and they arrive with a conversion problem that traditional optimization frameworks were never designed to solve. These products, built rapidly with AI assistance by solo founders or small teams, frequently launch with real traction. The build-fast dynamic means they can reach users within hours of conception. What they almost never have is the analytics instrumentation, attribution setup, or funnel documentation that a traditionally developed SaaS product accumulates over months of deliberate growth infrastructure work. With 2.4 million new apps created via vibe-coding platforms in the past 12 months alone and 85% of new greenfield projects starting AI-first, this is not a niche edge case. It is an endemic structural gap affecting an enormous and growing segment of the SaaS market.

The absence of a dedicated growth team compounds the problem in a specific way. Without baseline funnel data, ad-hoc optimization efforts have nothing to anchor against. A founder who notices declining trial signups cannot determine whether the issue lives in top-of-funnel traffic quality, landing page messaging, onboarding friction, or activation failure. Every lever looks equally plausible because no stage has been measured. This is categorically different from the optimization gaps discussed earlier in this analysis, where data exists but is misread. For vibe-coded apps, the data pipeline itself is missing from day one.

The onboarding-to-activation gap is where this plays out most damagingly. The vibe-coding workflow is oriented around shipping a feature that works, not designing a user journey that converts. There is no natural step in the process that prompts a builder to define an activation event, map the path to the first value moment, or instrument dropout between signup and engagement. The result is that high trial abandonment gets misread as a traffic volume problem when the actual failure is an undesigned activation path.

FunnelKeeper is built specifically for this profile. It provides out-of-the-box funnel dashboards, attribution tracking, and user adoption visibility that gives vibe-coded app builders the conversion infrastructure of a Series B growth team without requiring one. Rather than retrofitting analytics as a later-stage project, FunnelKeeper makes funnel visibility the starting condition, so every week of traction generates compounding conversion intelligence rather than compounding blind spots.

The Mobile Conversion Gap: SaaS's Most Underoptimized Lever

Mobile devices now account for 82.9% of all landing page visits, yet desktop still converts approximately 8% better. For SaaS sign-up flows already operating at a median conversion rate of just 3.8%, that gap is not a minor technical footnote; it is a measurable, recurring revenue leak that grows proportionally with every campaign you run.

The mobile conversion penalty has three primary drivers. First, form friction compounds on smaller screens: three-field forms convert at 10.1% while nine-field equivalents drop to 3.6%, and 81% of users abandon forms they have already started. Poor autofill support and missing OAuth sign-in options make every additional field disproportionately expensive on mobile. Second, load speed functions as a direct revenue variable; every one-second delay in page load beyond 2.5 seconds costs 7% in conversions, and mobile networks routinely expose SaaS pages to this penalty. Third, CTAs built for cursor precision fail on touchscreens. Buttons sized below the 44x44 pixel minimum recommended by Apple and Google, or positioned outside the natural thumb zone in the bottom-center of the screen, suppress tap rates regardless of how strong the copy is.

The highest-ROI interventions are consistently undertreated. Progressive disclosure forms that collect only an email address on the first step, then surface role or company questions conditionally on step two, reduce cognitive load while preserving qualification data. Multi-step forms outperform single-page equivalents by 21%. Mobile-isolated A/B tests on headline copy, CTA placement, and form length are particularly valuable because running device-blended tests averages out the mobile signal entirely.

None of this is actionable until you instrument your funnel dashboard to report mobile and desktop conversion rates as separate metrics. Blended averages absorb the mobile penalty invisibly. Segmenting by device in your analytics is the diagnostic step that converts a hidden leak into a quantifiable, addressable optimization target.

How to Build a Documented Conversion Optimization Strategy in 5 Steps

The five steps below convert optimization from an abstract intention into a repeatable operational system. Each step builds on the previous one, which means skipping any step undermines the entire chain.

Step 1: Instrument Your Funnel Before Touching Anything Else

Every optimization decision you make is only as reliable as the measurement infrastructure underneath it. Define your funnel stages explicitly: visitor, lead, trial signup, onboarding completion, and paid activation. Then confirm that each transition carries a tracked event with a reliable data source behind it. This is the step the majority of teams skip, and it explains why optimization efforts so frequently produce inconclusive results. Without clean instrumentation, you are testing hypotheses against noisy data and drawing conclusions that cannot be replicated.

Step 2: Establish Baselines and Calibrate Against 2026 Benchmarks

Once your tracking is solid, document your current stage-level conversion rates before running a single test. Then compare those rates against 2026 benchmarks. The visitor-to-lead gap between average performers at 1.5% and elite performers at 8 to 15% is not an abstract reference point; it is a precise revenue gap you can calculate in ARR. Stage-level benchmarking turns optimization from a vague improvement goal into a specific number you are closing toward. McKinsey data confirms that companies pursuing systematic funnel optimization achieve 30 to 50% conversion rate improvements, and that outcome begins with knowing exactly where you currently stand.

Step 3: Rank Leaks by Volume, Not by Rate

This is where most analytical approaches go wrong. A 20% drop-off rate at a stage receiving 10,000 monthly visitors represents 2,000 lost opportunities. A 50% drop-off rate at a stage receiving 100 visitors represents 50. Fix the larger volume problem first, every time. Mid-funnel stages, particularly the transition from qualified lead to active trial, account for approximately 60% of revenue loss in growth-stage SaaS companies precisely because they carry high traffic volume but receive disproportionately low optimization attention.

Step 4: Run Attribution-First Channel Analysis Before Testing Creative

Multi-touch attribution tells you which channels are actually contributing to conversions across the full journey and which are consuming budget at touchpoints that never close. With B2B sales cycles now averaging 6.5 months and buying committees involving roughly 13 stakeholders, last-touch attribution systematically misdirects budget. Reallocate spend based on attribution data before launching new creative tests; otherwise you are optimizing messaging on a channel mix that was already broken.

Step 5: Formalize a Testing Cadence and Document Every Finding

Conversion optimization compounds when each test informs the next hypothesis. A documented testing log captures the hypothesis, the variant tested, the sample size reached, the result, and the planned next iteration. This structure transforms individual wins into institutional knowledge. Companies that excel at systematic lead nurturing generate 50% more sales-ready leads at 33% lower cost, and that outcome requires a documented process, not a series of disconnected campaigns.

Conversion Optimizer Benchmarks: How Does Your Funnel Stack Up?

Benchmarks only create value when you know which number to compare yours against and what the gap is actually telling you.

The visitor-to-lead conversion rate is the single most diagnostic metric at the top of your funnel. The average B2B SaaS company converts 1.5% of website visitors into leads. The top 10% convert between 8% and 15%. That is not a marginal performance difference; it is a structural one, and it compounds directly into ARR. If your rate sits below 3%, the diagnosis almost always points to one of three root causes: your top-of-funnel messaging is not matching buyer intent, your incoming traffic quality is too broad, or your lead capture mechanism creates unnecessary friction before the visitor sees enough value to act.

Channel-level benchmarks add a second layer of context. Organic search drives the majority of SaaS buyer journeys, making it the highest-volume entry point in the funnel even where individual session conversion rates trail other channels. Software review site traffic, by contrast, arrives with purchase intent already established, which is why it consistently converts at 5% to 7%. Email nurture leads convert at significantly higher rates than cold traffic because the relationship and trust have been built before the conversion moment arrives. Understanding which channel is delivering volume versus which is delivering conversion efficiency is essential for allocating optimization resources correctly.

The cross-industry average conversion rate provides a useful systemic floor. A SaaS company consistently converting below the broad cross-industry median is not facing a marginal optimization opportunity; it is facing a structural funnel problem in messaging, targeting, or funnel architecture that incremental testing will not solve.

Trial-to-paid benchmarks vary sharply by pricing model. Free trials with credit card requirements routinely convert at 40% to 60%. Fourteen-day opt-in trials average 15% to 25%. Freemium and demo-led models operate in entirely different ranges. The companies that achieve the most significant conversion lifts share one documented habit: they measure and report this stage explicitly as a marketing metric, not just a product metric.

Tracking your rates against these benchmarks inside a live funnel dashboard, updated continuously rather than reviewed quarterly, is what converts benchmark data from reference material into an active decision-making tool.

Closing the Conversion Gap: Where to Start

The conversion performance gap between top SaaS companies and average performers is not a creative problem or a budget problem. It is a visibility and systems problem, and that distinction matters because it is entirely solvable with the right instrumentation in place. Companies converting at 8 to 15% are not running fundamentally different marketing than companies stuck at 1.5%. They are operating with better data, tighter feedback loops, and documented processes that surface leakage before it compounds into ARR loss.

For the 68% of B2B SaaS companies currently operating without a documented funnel strategy, the single highest-leverage first action is funnel instrumentation. Every optimization tactic covered in this guide, from activation improvements to mobile UX fixes, depends on reliable stage-level data as a prerequisite. You cannot fix what you cannot see, and without baseline metrics at each funnel stage, every optimization decision is a guess dressed up as a strategy.

Attribution clarity must come before channel scaling. Misallocated spend suppresses conversion rates silently, inflating CAC while masking which channels are actually producing pipeline. Fixing attribution consistently delivers faster conversion gains than any individual landing page test because it reallocates existing budget toward already-performing channels.

FunnelKeeper provides the funnel dashboards, attribution intelligence, and user adoption tracking that SaaS teams and vibe-coded app builders need to move from optimizing blindly to optimizing systematically. Start with your biggest leak, document your baseline, and complete your first attribution audit within the next 30 days. Those three actions alone will position you ahead of the majority of the market.

The SaaS Conversion Optimizer: Why 68% of Companies Are Optimizing Blindly