SaaS Sales Process: PLG vs SLG vs Hybrid — When to Use Each

Every B2B SaaS founder I talk to between $1M and $15M ARR eventually asks me the same question, usually phrased some variation of "should we go more product-led or more sales-led?" The framing is wrong. PLG and SLG are not strategic identities you pick once and stick to — they are operating motions that should follow the shape of your revenue, not lead it.

The single variable that determines which motion fits is ACV — annual contract value — combined with deal complexity. Below roughly $5K ACV, the unit economics of a human-led sales process collapse: a 30-minute discovery call costs more than the deal's first-year margin contribution. Above roughly $25K ACV, pure self-serve also collapses, but for the opposite reason — buyers at that price band expect a conversation, multiple stakeholders touch the decision, and procurement enters the picture whether you want it there or not. The interesting territory, and where most B2B SaaS lives, is the $5K–$25K band where neither pure motion works and the question becomes how to architect a hybrid that doesn't trip over itself.

This article is opinionated. Pure PLG is rare in B2B above $25K ACV — most companies that claim it are actually running hybrid with a marketing department that prefers the PLG label. Pure SLG below $5K ACV burns cash on discovery calls that produce $400 monthly contracts. The work is figuring out which motion fits your current revenue band, knowing when the band shifts and the motion has to evolve with it, and engineering the hybrid in a way where PQLs and MQLs each get the SLA they deserve. The architecture of the underlying repeatable sales process — stages, exit criteria, SLAs — is the same in both motions. What changes is who runs which stage, how the buyer enters the funnel, and where the human touch lands.

Before designing the motion, you have to honestly diagnose which territory your product and your buyer sit in. Four signals matter, in this order.

Signal 1 — ACV band. The defining variable. Below $5K ACV per year, your unit economics will not support a human-touched sales process unless every touch is highly automated. Between $5K and $25K, you can support an AE-led close but not an AE-led top-of-funnel — too expensive to qualify cold. Between $25K and $100K, you need an AE-led process end to end, with marketing feeding qualified pipeline, but the deal is still small enough that PLG-style self-serve trials can pre-qualify buyers before sales engages. Above $100K ACV, you're in classical SLG territory: every meaningful deal is human-led from the first conversation.

Signal 2 — Time to value. How long between a buyer hitting your product and them experiencing something useful? If a user can configure your product and get value in under 30 minutes without a human guiding them, PLG motion mechanics work — the product itself sells. If meaningful value requires implementation, integration, change management, or training on the buyer side, no amount of UX polish makes self-serve viable. Even a $4K ACV product with a 90-day implementation needs SLG-style hand-holding.

Signal 3 — Buyer count per deal. Single-buyer deals (one person decides, one person pays, one person uses) can run self-serve at almost any ACV up to about $15K. Multi-buyer deals (a champion, an economic buyer, an end user, a procurement gate) require human orchestration the moment more than two distinct stakeholders are involved, regardless of ACV. The B2B SaaS reality is that buyer count climbs with company size — a $50/user/month product becomes a $30K deal at a 500-person company, and the buying committee makes it SLG even though the unit price screams PLG.

Signal 4 — Sales cycle length. If your typical closed-won deal closes in under 14 days from first touch, you're already in PLG territory whether you call it that or not — humans cannot meaningfully intervene at that speed in volume. If your typical cycle runs 45–90 days, you're SLG, and the process needs to support multithreading, multiple stakeholder calls, and asynchronous decision waiting periods. If cycles split into a bimodal distribution — a chunk closing in 7–14 days and another chunk taking 60–90 — that's the strongest signal you should be running hybrid, because you have two distinct buyer journeys and one process is currently mistreating one of them.

The diagnostic exercise is simple but founders rarely do it cleanly. Pull your last 50 closed-won deals. Plot ACV against sales-cycle length. Count distinct stakeholders per deal. If the cloud you see is tight and below $5K ACV with sub-14-day cycles, you're PLG. If it's tight and above $25K with 60-day cycles, you're SLG. If the cloud is wide and bimodal — which I see more often than anything else — you're already hybrid, and the question is whether you're hybrid-by-design or hybrid-by-accident.

The stages are the same. What happens inside each stage is radically different. Here's how the playbook diverges across the four moments that matter — discovery, demo or POC, pricing conversation, and close.

Discovery. In SLG, discovery is a 30–45 minute structured conversation where the AE maps the buyer's problem, timeline, budget owner, and decision process. Twelve discovery questions, scripted, the rep listens 70% of the time. In PLG, discovery happens inside the product — usage patterns reveal the use case, configuration choices reveal the team structure, the upgrade trigger reveals the buying intent. The "discovery call" in PLG is usually a 20-minute call after the user has already used the product, where the rep validates what the data already says rather than uncovering anything new. If you find yourself running 45-minute discovery calls with users who already use your product daily, your motion is misaligned.

Demo or POC. SLG demos are tailored to the buyer's stated problem, run by the AE or a sales engineer, and serve to validate solution fit against the success criteria articulated in discovery. POCs in SLG are scoped, time-bounded (typically 14–30 days), and have explicit success criteria the buyer agrees to before the POC starts — otherwise the POC becomes free consulting. PLG skips the demo entirely. The product is the POC, every day, with no time bound. The PLG equivalent of a POC failing is the user not upgrading — which is harder to coach against because there's no live conversation in which to surface objections.

Pricing conversation. SLG pricing is negotiated. The list price is a starting point, discount tiers are pre-approved, and the AE has authority to flex on terms (payment cadence, multi-year, expansion commitments) in exchange for protected commercials. The conversation happens live, on a call, with the economic buyer in the room. PLG pricing is published. The list price is the price. Volume tiers exist, but they're rule-based and visible. The pricing conversation in PLG is asynchronous — the buyer compares your published tiers to alternatives without you in the room. If you're in a hybrid motion and you let your AE start discounting in a deal that came in through self-serve, you've just trained that customer to never trust your published pricing again, and they'll bring that expectation to their renewal.

Close. SLG closes happen on a call, with a signed paper agreement, after a mutual close plan was set 14–30 days earlier. The AE drove the timeline. PLG closes happen when the buyer clicks an upgrade button, sometimes at 11pm on a Tuesday, with no human present. The PLG "close plan" is product mechanics: pricing-page visit triggers, expansion limits hit, trial expiring. The reason hybrid is hard is that mid-market deals often start PLG (self-serve trial) and finish SLG (a procurement-led negotiation), and the rep who picks up the deal at the SLG end has to honour the pricing the buyer saw on the PLG side. Bait-and-switch pricing on a deal that started self-serve is the fastest way to lose hybrid deals at the procurement gate. The pricing logic of where this breaks is partly covered in the sales playbook template for B2B SaaS, which addresses the segment-by-ACV pricing question directly.

Hybrid motion is where most B2B SaaS lives between $5K and $50K ACV, and it's where most teams operate with a half-built system. The mechanics that separate working hybrid from accidental hybrid sit in three places: how a PQL is defined, when an AE is allowed to engage, and which buyers are deliberately left in the self-serve lane.

PQL definition (binary, like MQL but stricter). A product-qualified lead is a self-serve user whose usage pattern indicates buying intent for a paid plan one or more tiers above their current state. The trigger is composite, not single-metric. Examples of PQL composites I've installed:

• Free user who has invited 3+ teammates and used the core feature 5+ times in the last 14 days, plus their company matches ICP firmographics (≥50 employees, target industry).
• Trial user who has connected 2+ integrations and has at least 2 active users from the same email domain in week 1 of trial.
• Existing paid customer whose usage has exceeded the soft limit of their current plan by ≥40% for three consecutive billing periods.

The key design decision: PQL must require both behavioural depth (they're using it for real) and firmographic fit (they're someone you'd actually want to sell to). Behavioural-only PQLs flood the AE queue with hobbyists who happen to be enthusiastic. Firmographic-only PQLs are just MQLs with extra steps.

AE engagement triggers and SLAs. The PQL SLA must be different from the MQL SLA, because the buyer state is different. An MQL is someone who showed inbound intent — they want a conversation. A PQL is someone using your product — they may or may not want a conversation. Default rule I install: PQL gets a personalised, low-friction message within 4 business hours of trigger fire. The message is short, references specific product usage ("I see you've connected your Salesforce and have your team set up — happy to help if you're thinking about how to scale this across more teams"), and offers a 15-minute call. Aggressive demo-pitch outreach on PQL signups kills the trust the product built — it makes the user feel surveilled.

If the PQL does not respond, the AE backs off after two touches. The lead returns to a marketing nurture stream that matches the product behaviour pattern. Critically: PQLs that haven't converted to a sales conversation but are still using the product do not get re-engaged by sales until a new product-side trigger fires. The product is doing the selling — let it.

The self-serve lane (and why it exists deliberately). Not every PQL should get an AE. A self-serve user converting to a $4K ACV paid plan does not need a sales conversation — the AE cost (let's call it $300 per touched lead in time and overhead) is 7.5% of the first-year contract value. Worse than that, the AE conversation often slows down the conversion, because the AE wants to qualify and the buyer wanted to click upgrade. I usually define a deliberate self-serve floor: below a certain ACV expansion threshold, the AE does not engage on the upgrade itself, only on the expansion that comes after. The threshold is typically the price point where AE involvement adds enough deal-size expansion to justify the cost — for most B2B SaaS, that lands somewhere between $8K and $15K ACV.

The hybrid model has a clean operational test: if you compare closed-won rate and time-to-close for AE-touched PQLs vs untouched self-serve conversions at the same ACV band, the AE-touched cohort should win on either rate, deal size, or both — meaningfully. If the AE-touched cohort wins on neither, the AE is destroying value, not adding it, and the threshold needs to move up. The pipeline coverage benchmarks for hybrid teams reflect this — coverage ratios that look healthy in pure SLG often look anaemic in hybrid because the self-serve lane is producing closed revenue without ever appearing in the pipeline.

Hybrid motion failures come in two predictable shapes, and they correspond to which side of the org has more political weight.

Failure mode 1 — Sales force-closing self-serve signups. The sales team is on quota. They see a flood of self-serve signups, much of which converts to paid without their involvement, and quietly resent that revenue isn't accruing to their book. They start prospecting into the signup base, treating PQLs (and worse, all signups) as MQLs requiring outbound qualification. Within two quarters, every self-serve signup is getting a demo-pitch email within 24 hours of trial start. The signups stop converting at the self-serve rate they used to, the AE-led cohort numbers look great in isolation, but total bookings flatten or decline. The product team can see it in the data; sales leadership won't acknowledge it because the AE-touched numbers look strong.

The fix is an explicit no-touch policy for sub-threshold signups, enforced by routing rules in the CRM, not by AE discipline. AE compensation must also exclude self-serve conversions below the threshold — if reps can earn quota credit for a self-serve $400/month upgrade they touched once, they'll touch every one of them.

Failure mode 2 — Product starving the sales motion of leads. The opposite failure, equally common in product-led companies with a smaller sales team. The product team optimises hard for self-serve conversion, removing friction at every step. They publish pricing low to maximise top-of-funnel volume. Self-serve numbers look great. Then a mid-market or enterprise buyer comes in, lands on the same pricing page, sees the published $99/month tier, and decides the product isn't serious enough for their use case. They leave without contacting sales. Sales sees no inbound pipeline from above the threshold; product sees no funnel issue because the lower-tier numbers are strong.

The fix is a deliberate "talk to sales" path for above-threshold use cases — published team-tier pricing that says "contact us for >50 users," enterprise-tier features behind a sales-led gate, custom pricing language above a certain volume. The pricing page becomes the segmentation device. Companies running pure published pricing across all tiers up to enterprise systematically lose the deals that would have justified having an SLG motion in the first place.

The pattern under both failure modes is the same: the org has not decided whose accountability owns which deals at which ACV. Product owns conversion up to the threshold; sales owns conversion above it. The boundary is the threshold, it is mechanical, and disputes about whose deal a particular signup is must be resolved by ACV band, not by who touched it first. If you want a deeper view of how the broader transformation engagement establishes this kind of boundary, sales transformation consulting engagement walks through the org-design side.

Here's the matrix I use when a founder asks me "which motion should we run." It is opinionated and it errs on the side of clarity over nuance, because most founders are trying to fix being indecisive, not being too decisive.

                    │ Single-buyer  │ 2-3 buyers      │ 4+ buyers
                    │ <14 day cycle │ 30-60 day cycle │ 60-120 day cycle
────────────────────┼───────────────┼─────────────────┼─────────────────
ACV  <$5K           │   Pure PLG    │   Pure PLG      │   Reconsider ICP
                    │               │   (rare)        │   (deal too
                    │               │                 │   expensive to
                    │               │                 │   serve)
────────────────────┼───────────────┼─────────────────┼─────────────────
ACV  $5K–$15K       │   PLG + PQL   │   Hybrid        │   Hybrid leaning
                    │   sales lane  │   (PQL→AE on    │   SLG
                    │               │   threshold)    │
────────────────────┼───────────────┼─────────────────┼─────────────────
ACV  $15K–$50K      │   Hybrid      │   Hybrid leaning│   SLG
                    │   (rare —     │   SLG           │
                    │   investigate │                 │
                    │   why low     │                 │
                    │   buyer ct)   │                 │
────────────────────┼───────────────┼─────────────────┼─────────────────
ACV  $50K–$200K     │   SLG         │   SLG           │   SLG with
                    │   (very rare) │                 │   enterprise
                    │               │                 │   playbook
────────────────────┼───────────────┼─────────────────┼─────────────────
ACV  >$200K         │   SLG         │   SLG           │   SLG with
                    │   (very rare) │                 │   procurement &
                    │               │                 │   security path

Three decisions follow from the matrix.

If the bulk of your revenue sits in one cell, run the motion that cell prescribes. Do not run an SLG team because you aspire to enterprise — run it when you have enterprise deals. Hire AEs to close deals that exist today, not deals you want to exist next year.

If your revenue sits across two adjacent cells, hybrid is correct, and the threshold between PLG and SLG handling is the ACV at the boundary between those cells. Build the PQL routing rules to fire at that threshold.

If your revenue spans three or more cells, you have a segmentation problem first and a motion problem second. The most common version of this is a company with $400/month self-serve users alongside $80K enterprise deals — two completely different products being sold to two completely different buyers under one P&L. The motion question can't be answered until the segmentation is named honestly, sometimes as separate go-to-market motions inside one company with deliberate operational independence.

Motion choice is not permanent. Most B2B SaaS companies cycle through this matrix as they grow — PLG at seed, hybrid in Series A, SLG-dominant by Series B with PLG retained for SMB self-serve. The trap is committing to a motion label out of identity rather than from the revenue band you're actually in this quarter. Pull your last 50 deals, place them on the matrix, and design the motion around where the deals actually live. If you want the operational design that sits underneath whichever motion you land on, the project-based transformation engagement walks through the install — process, playbook, instrumentation — across the full 12–18 week implementation. The motion question is upstream of the install; the install is what makes the chosen motion actually work in practice.