How'd you fix Degreed's revenue issues in 2026?

The two options on the table

By 2026 Degreed faces a genuine fork, and the tempting answer — do both at full intensity — is the one that burns the balance sheet. The learning experience platform category Degreed helped create has been squeezed from two directions. Workday Learning arrives free inside an HCM suite the customer already bought. Cornerstone OnDemand absorbed EdCast and folded LXP functionality into a broader talent cloud. When a capability ships bundled at zero marginal cost, a standalone vendor charging six figures for the same capability is not a competitor, it is a line item waiting to be cut at renewal. Meanwhile general-purpose AI assistants will draft a passable skills-to-role mapping on request, which erodes the perceived scarcity of a skills taxonomy sold as a static artifact. Degreed's free skills-discovery tool generated awareness but converted a small single-digit percentage into enterprise contracts, so top-of-funnel volume never became revenue. The company sat in the classic middle: too expensive to win on platform features against a suite, too horizontal to win the skills-intelligence category against a technical vertical player or an outcome-contract education vendor.

Option A — outcome-locked skills intelligence, sold direct. Degreed stops selling seats and starts selling a measured result: a documented shift of hiring from external requisitions to internal transfers, with the fee anchored to the savings. Contracts run roughly $70K to $300K per year for mid-market organizations in the $100M to $1B revenue band. The buyer is a Chief Talent Officer or VP of Learning who has been asked by a CFO to defend the learning budget and cannot do it with course-completion counts. The deliverable is a quarterly business review showing time-to-fill for internally filled roles, cost-per-hire delta between internal and external fills, and 12-month retention of internally placed employees versus external hires. Degreed's defensible input is longitudinal: assessment results, project completions, manager ratings, and peer endorsements accumulated across years of enterprise use. A public model has none of that history for a specific employer's workforce.

Option B — vertical bundles distributed through channel partners. Degreed stops paying for its own field sales in the mid-market and instead embeds the skills-intelligence engine as a white-label or co-branded module inside payroll, HRIS, and benefits platforms — the Rippling, Gusto, BambooHR, Paylocity tier — where tens of thousands of mid-market employers already log in weekly. Degreed takes a revenue share in the 20% to 30% range on subscriptions sold through the partner. The bundles are verticalized for high-churn sectors — healthcare, logistics, customer success, frontline manufacturing — where annual turnover in critical roles runs high enough that a few points of improvement is a real number on the P&L. Pricing in the channel is deliberately smaller and simpler than the direct motion, because a partner's rep will not carry a nine-month enterprise sales cycle.

The trade-off is not quality of idea, it is cash-cycle length versus margin per logo. Option A produces large, defensible, high-margin contracts on a slow clock: six to nine months of enterprise cycle, a fully loaded customer acquisition cost in the $25K to $40K range per closed deal, and a real risk that the outcome clause makes revenue recognition lumpy. Option B produces small, fast, lower-margin logos: acquisition cost per deal potentially under $5K because the partner's rep does the selling, but Degreed surrenders a fifth to a third of the subscription and loses direct control of the customer relationship, the renewal conversation, and the upsell path. A RevOps team modeling this should not treat them as competing philosophies — they are different segments of the same market with different unit economics, and the sequencing question is which one funds the other.

There is a third temptation worth naming and rejecting: keep selling the horizontal LXP at a discount to defend logo count. That path trades away the only asset that is not commoditized. Every dollar of discount taken to match a bundled suite trains the market that Degreed's skills layer is worth roughly what Workday charges for it, which is nothing incremental. Discounting is a legitimate retention tactic on individual renewals; it is not a revenue strategy.

How to decide between them

The decision is not a coin flip and it is not a taste question — it resolves against three testable conditions in the existing book of business. Run the test on real account data before committing headcount.

Condition one: does the installed base already produce mobility evidence? Pull the last four quarters of accounts and ask how many can produce, from data Degreed already holds, a defensible count of internal transfers where a Degreed skill signal was in the path. If a meaningful slice of accounts can — call it a fifth or more, and the count is auditable rather than anecdotal — the outcome-locked motion has proof to sell with, and the direct path is viable now. If almost no account can produce that evidence, an outcome contract is a promise against an unmeasured process, and signing it transfers delivery risk onto Degreed with no operational history to price it. In that case the honest sequence is: instrument first, sell outcomes second.

Condition two: what is the current fully loaded CAC payback? If direct-sold contracts at $70K to $300K are paying back acquisition cost inside roughly 12 to 18 months on a gross-margin basis, the direct motion is self-funding and deserves more capital. If payback stretches past two years — which it will if the cycle is nine months and win rates are thin — direct selling is consuming cash the company does not have, and channel becomes the survival move regardless of its margin haircut.

Condition three: is there a partner willing to co-sell, not just co-list? A logo on a partner marketplace page is worth approximately nothing. The condition that matters is whether a partner will put the module in front of its own installed base with its own sequence — in-product placement, a lifecycle email, a named partner-success owner. If no partner will commit to distribution mechanics in writing during a pilot, the channel number is a spreadsheet fantasy and the plan should not be built on it.

The decision tree resolves to a default for most realistic states of the business: instrument the mobility data, run outcome-locked contracts in the segment where evidence already exists, and pilot channel in parallel at deliberately small scale — three to five partners, not fifteen. Fifteen partners is not a channel program, it is fifteen partial integrations nobody owns.

One more decision input belongs in the RevOps model: churn concentration. If a large share of at-risk ARR sits in accounts that bought Degreed as a content-aggregation layer, no repositioning saves them — those accounts churn to the bundled suite, and the plan should assume it rather than budget for a save. Modeling a 2026 recovery on retaining commodity-LXP buyers is how a turnaround plan becomes fiction in month five.

The numbers behind each option

Numbers here are illustrative model inputs built from publicly discussed hiring-cost ranges, not reported Degreed financials. Treat them as the arithmetic a RevOps team should run with its own actuals substituted in.

Direct outcome-locked contracts. The economic engine is the spread between external and internal hiring cost. External hires commonly cost an employer roughly $4,000 to $20,000 all-in depending on seniority — agency or sourcing spend, recruiter time, assessment, onboarding drag. Filling the same role internally typically costs far less, in the $500 to $2,000 range, because sourcing is nearly free and ramp is shorter for someone who already knows the systems and the people. Take an organization making 200 hires a year. Shifting 30% of them — 60 hires — from external to internal produces a per-hire saving somewhere between $2,000 (low external, high internal) and $19,500 (high external, low internal). Across 60 hires that is roughly $120,000 at the pessimistic end and about $1.17 million at the optimistic end. A realistic midpoint using a $10,000 external and $1,200 internal assumption lands near $528,000.

That range sets the price ceiling honestly. A $70K contract against $120K of savings is a hard sell — the buyer sees a thin margin and a lot of change management. A $70K contract against $500K of savings closes, because the ratio survives a skeptical CFO discounting it by half. So the segmentation rule falls out of the arithmetic: outcome-locked pricing works where hiring volume is high enough or roles are senior enough that the savings pool clears roughly five to seven times the contract value. Below that ratio, sell a flat subscription and stop pretending the outcome clause is the value driver.

Vertical bundles. High-churn sectors are the target because the savings pool is turnover, not just hiring cost. Healthcare, logistics, and frontline customer roles routinely run turnover well above the cross-industry average, and the replacement cost of a departing employee is a meaningful fraction of annual salary once you count vacancy coverage, overtime, training, and productivity ramp. A vertical bundle priced in a $15K to $90K annual band per organization — sized to headcount and role criticality — has to demonstrate a turnover reduction in the low single digits of percentage points to pay back inside two to three quarters. That payback claim is the entire sales argument, so it must be measured on the customer's own attrition data, not a benchmark deck.

Channel economics. The channel case is a CAC argument, not an ACV argument. Direct mid-market acquisition at $25K to $40K per deal against a $40K annual contract means the first year is roughly a wash before delivery cost. Channel acquisition under $5K per deal against the same $40K contract, minus a 25% partner share, nets Degreed $30K in year one on $5K of acquisition. The channel deal is worth less gross and more net. Model a pilot honestly: three to five partners, and a realistic year-one target measured in low hundreds of logos, not thousands. At 200 logos averaging $40K, gross channel-attributed revenue is $8 million; after a 25% partner share Degreed nets roughly $6 million, against pilot integration and partner-enablement cost that is real and should be budgeted at engineering-quarters, not zero.

The marketplace and data add-ons — where to be conservative. Two adjacent revenue ideas get proposed in every plan like this and both deserve skepticism in year one. A two-sided skills-based hiring marketplace, where employers pay for access to verified candidate profiles from other Degreed clients, is a genuine network-effect asset — verified assessment results and manager ratings are strictly better signal than self-reported profiles on a job board. But it is also a cold-start problem wrapped in a consent problem: employees must opt into external visibility, and employers must accept that their own talent is discoverable. Budget it as an experiment with a small design-partner cohort, not as a revenue line in the 2026 plan.

The second is a skills-depreciation data product — a subscription that scores how fast a given skill is losing market value, derived from job-posting trends, wage movement, and technology adoption curves. The margin profile is genuinely attractive because it is analysis rather than content production, and quarterly refreshes make renewal structural rather than optional. Priced as a $15K to $40K annual add-on onto an existing subscription, it needs no new acquisition cost — the account manager presents it inside the QBR that is already scheduled. But it competes against established labor-market data providers, and Degreed's differentiated angle is narrow: cross-validating external market signal against a customer's own internal skill inventory. Sell that specific angle or do not sell it.

What the blended model should look like. A defensible 2026 plan weights direct outcome contracts as the majority of new ARR because the margin funds everything else, channel as the logo-volume and CAC-efficiency engine, and add-ons as expansion revenue inside the installed base. If channel is projected to carry more than roughly a third of new ARR in its first year, the plan is assuming partner performance that has not been demonstrated.

Implementation sequence and the RevOps work underneath

The strategy fails on execution details more often than on positioning, and nearly all the execution risk sits in RevOps: instrumentation, contract mechanics, comp design, and partner operations.

Quarter one — instrument before you promise. No outcome contract should be signed before the measurement is real. That means a defined internal-transfer event in the customer's system of record, a documented attribution rule for when a Degreed signal counts as being in the path, a baseline period of at least the prior four quarters of the customer's own hiring data, and agreement in writing on the data source before the contract starts. The single most common failure in outcome-based pricing is a dispute at true-up because the two sides were counting different things. Write the counting rule into the order form.

Quarter one, in parallel — pick three to five channel partners and one vertical. Do not launch four verticals into five partners. Pick the vertical where the existing customer base already has reference logos and pick partners whose installed base skews to that vertical. The integration scope for a pilot should be deliberately shallow: authenticate, sync the employee roster and role data, surface the skills view in-product, and hand billing to the partner. Deep bidirectional sync is a quarter-three problem.

Quarter two — restructure comp and territories. This is where most repositionings quietly die. If the sales team is still compensated on total contract value with no modifier, no rep will invest in a channel-sourced $40K deal or accept the delivery risk of an outcome clause. Practical adjustments: pay accelerators on outcome-locked contracts that include a measured baseline, pay a reduced but real rate on channel-sourced revenue so partner-managers are not fighting the field for credit, and split territories so the same account is never worked by both a direct rep and a partner rep — conflict rules must be written before the first partner deal, not after the first escalation.

Quarter two to three — build the QBR as a product, not a slide. The outcome motion is only defensible if the quarterly review runs on live data the customer can audit. That means a standing dashboard with four measures: internal fill rate, time-to-fill for internal versus external, cost-per-hire delta using the customer's own cost assumptions rather than a benchmark, and 12-month retention of internally placed employees. Let the customer edit the cost assumptions themselves. A number the buyer typed in is a number they will defend to their CFO; a number from a vendor benchmark is one they will discount.

Quarter three — expansion motion inside the installed base. Only after the QBR dashboard is live does the data add-on become sellable, because the add-on's pitch depends on the customer already trusting the underlying skill inventory. Sequence matters: sell the measurement, earn the trust, then sell the analysis layer on top.

Quarter four — decide on the marketplace. By Q4 the pilot data will show whether employees opt into external visibility at any meaningful rate. If opt-in is negligible, kill the marketplace concept publicly and reallocate — a half-committed marketplace consumes engineering capacity indefinitely while producing neither revenue nor a decision.

Partner and vendor relationships, stated accurately. Placement-fee economics through staffing and RPO relationships are a legitimate adjacent revenue idea — a referral or placement fee on a successful hire, typically a percentage of first-year salary — but no specific partnership should be represented as existing unless it is signed. The same discipline applies to competitive framing: the relevant comparison set includes suite vendors like Workday and Cornerstone OnDemand, technical-skills specialists, outcome-oriented education providers, and the collaborative-learning platform 360Learning. Getting a competitor's name wrong in a board deck is a small error that costs disproportionate credibility.

The RevOps guardrails that keep this honest. Three measures should be reported internally every month regardless of what the sales narrative says: net revenue retention split between commodity-LXP buyers and outcome-contract buyers, so the mix shift is visible rather than hidden inside a blended number; CAC payback by motion, direct versus channel, computed on gross margin not revenue; and the true-up dispute rate on outcome contracts, which is the leading indicator of whether the measurement design actually works. If dispute rate climbs above a low single-digit share of contracts, the counting rule is broken and the fix is contractual, not commercial.

What to stop doing. Discounting the horizontal LXP to hold logo count. Quoting savings figures that cannot be reproduced from the customer's own data. Counting a partner marketplace listing as channel distribution. Forecasting marketplace revenue before opt-in behavior is observed. Each of these makes a quarter look better and makes the year worse, which is the specific failure mode that turned Degreed's positioning problem into a revenue problem in the first place.

Related questions

How does internal mobility actually reduce hiring cost?

Internal fills skip external sourcing spend and shorten ramp because the employee already knows the systems and stakeholders. Using $4,000–$20,000 external versus $500–$2,000 internal, shifting 60 of 200 annual hires internally saves roughly $120,000 to $1.17 million.

Which verticals justify a vertical bundle?

Sectors with structurally high turnover in critical roles — healthcare clinical staff, logistics supervisors, frontline customer success. There, a few points of retention improvement pays back a $15K–$90K annual contract in two to three quarters, measured on the customer's own attrition data.

Why is channel cheaper than direct for mid-market?

The partner's existing relationship replaces Degreed's outbound cost, dropping acquisition cost per deal from roughly $25K–$40K toward under $5K. The trade is a 20%–30% revenue share plus reduced control over renewal and upsell conversations.

What makes longitudinal skill data defensible against general AI models?

A public model can describe skills for a role generically. It cannot see a specific employer's years of assessment results, project completions, manager ratings, and peer endorsements — that accumulated internal history is what supports outcome-based pricing.

Should the hiring marketplace be in the 2026 plan?

Budget it as an experiment, not a revenue line. It has a real cold-start problem and depends on employees opting into external visibility. Read opt-in rates from a design-partner cohort before forecasting anything.

FAQ

What is the core positioning problem behind Degreed's revenue issues?

Standalone LXP capability became table stakes once suite vendors bundled it — Workday Learning inside the HCM suite, Cornerstone OnDemand after absorbing EdCast. A standalone platform priced in six figures for a bundled capability becomes a renewal-cycle cut. The fix is repricing against a measurable business outcome rather than platform features.

How should an outcome-locked contract be structured so it does not blow up at true-up?

Define the internal-transfer event in the customer's system of record, agree the attribution rule for when a skill signal counts, baseline at least four prior quarters of the customer's hiring data, and write all of it into the order form before signing. Most outcome-pricing disputes are counting disputes, not value disputes.

What ratio of savings to contract value makes outcome pricing work?

As a working rule, the demonstrable savings pool should clear roughly five to seven times contract value, so the deal survives a CFO discounting it by half. Below that ratio, sell a flat subscription — the outcome clause adds delivery risk without adding persuasion.

How many channel partners should a pilot include?

Three to five, aimed at one vertical where reference logos already exist. Fifteen partners produces fifteen half-finished integrations nobody owns. The qualifying test is whether a partner commits to in-product placement and a named success owner, not whether they will add a marketplace listing.

What should RevOps report monthly to keep the turnaround honest?

Net revenue retention split between commodity-LXP buyers and outcome-contract buyers so mix shift stays visible; CAC payback by motion on gross margin; and the true-up dispute rate on outcome contracts, which is the leading indicator that the measurement design is failing.

Is a skills-depreciation data product worth building?

Possibly, as expansion revenue inside the installed base rather than a new acquisition motion — it sells inside a QBR that already exists and refreshes quarterly, which makes renewal structural. But it competes with established labor-market data providers, so the only differentiated pitch is cross-validating external market signal against the customer's internal skill inventory.

Sources

• https://www.shrm.org/topics-tools/research — SHRM research on hiring cost and turnover benchmarks
• https://www.bls.gov/jlt/ — U.S. Bureau of Labor Statistics Job Openings and Labor Turnover Survey
• https://hbr.org/topic/subject/talent-management — Harvard Business Review on talent management and internal mobility
• https://www.gartner.com/en/human-resources — Gartner HR research on learning and talent technology
• https://www.mckinsey.com/capabilities/people-and-organizational-performance/our-insights — McKinsey on workforce and reskilling economics
• https://www.workday.com/en-us/products/talent-management/learning.html — Workday Learning product positioning
• https://www.cornerstoneondemand.com/ — Cornerstone OnDemand talent and learning platform
• https://360learning.com/ — 360Learning collaborative learning platform
• https://degreed.com/ — Degreed platform and product documentation
• https://www.sec.gov/edgar/search/ — SEC EDGAR full-text search for public filings

Related on PULSE

• How'd you fix Aston Carter's revenue issues in 2026?
• How'd you fix CyberCoders's revenue issues in 2026?
• How'd you fix Creative Financial Staffing's revenue issues in 2026?
• How'd you fix LanceSoft's revenue issues in 2026?
• How'd you fix Goodwin Recruiting's revenue issues in 2026?