Bruce MacVarish, AI innovation and GTM professional, has spent his profession constructing and pricing enterprise software program merchandise. That is the primary in a sequence of visitor posts from Bruce on GTM technique within the age of AI.
Synthetic intelligence is altering what prospects are keen to pay for.
When software program primarily automated duties, the affordable unit of sale was the seat, the license, or the hour. As AI techniques tackle extra of the work itself, from qualifying results in resolving assist tickets to writing code, the extra sturdy unit of sale turns into the outcome these techniques produce. That is the shift from promoting functionality to promoting Outcomes.
This piece lays out the rules behind an outcomes-based worth creation technique, the go to market practices that assist it, and the measurement infrastructure required to make end result primarily based pricing credible and defensible. It closes with a set of working suggestions for leaders who need to construct or convert a enterprise round Outcomes fairly than utilization or effort.
The central argument is easy. Consequence primarily based promoting is simply as robust because the measurement layer behind it. And not using a rigorous, clear, and mutually trusted method to outline, attribute, and confirm outcomes, an Outcomes technique is only a advertising label on prime of a standard subscription. With that measurement layer in place, Outcomes grow to be a real supply of differentiation, pricing energy, and buyer loyalty.
Why Worth Creation Is Shifting to Outcomes
For a lot of the software program period, distributors bought entry. A buyer paid for a seat, a license, or a block of compute, after which needed to do the work of turning that entry into worth. The seller’s job ended at supply of the device. The client’s job started at adoption.
AI modifications this association as a result of it closes the hole between entry and outcome. A mannequin that drafts a contract, resolves a assist ticket, or qualifies a gross sales lead is just not merely a device that helps an individual do a process sooner. In lots of instances it does the duty. When the system itself produces the deliverable, charging for entry to the system fairly than for the deliverable begins to look arbitrary to the client. Why pay for a seat when no particular person wants to sit down in it?
That is the underlying drive pushing worth creation towards Outcomes. It’s not a advertising pattern a lot as a pure consequence of automation maturing to the purpose the place the system, not the consumer, is the first unit of manufacturing. Consumers more and more ask a easy query: what did I get, not what did I’ve entry to.
From Effort to End result
Conventional software program and providers pricing is anchored to inputs: hours billed, seats provisioned, tickets processed, tokens consumed. Consequence primarily based pricing is anchored to outputs: leads transformed, instances resolved, prices prevented, income generated. The excellence issues as a result of inputs are straightforward to measure and onerous to connect with worth, whereas outputs are onerous to measure and immediately characterize worth. An Outcomes technique accepts the tougher measurement drawback in trade for a pricing mannequin that’s naturally aligned with what the shopper really cares about.
Rules of an Outcomes Primarily based Technique
Precept One: Outline the Consequence Earlier than the Supply
An Consequence should be outlined in phrases the shopper already makes use of to run their enterprise, comparable to value per decision, certified pipeline generated, or days gross sales excellent. If the definition requires inventing a brand new metric that solely exists inside the seller’s dashboard, adoption will stall as a result of the client has no inner body of reference for judging success.
Precept Two: Value the End result, Not the Mechanism
The pricing dialog ought to heart on the Consequence delivered, with the underlying AI system handled because the mechanism that makes the pricing attainable fairly than because the product itself. This doesn’t imply hiding the expertise. It means holding the business dialog centered on the enterprise outcome, whereas the technical dialog, which nonetheless issues to technical patrons, occurs individually.
Case Research: Intercom Fin
Intercom fees a small charge solely when its Fin agent totally resolves a buyer dialog, with no cost for failed makes an attempt, and a decision is counted when the shopper confirms it or doesn’t comply with up. The Consequence is outlined in a metric assist leaders already observe, and the seller absorbs the price of failed makes an attempt fairly than passing that threat to the shopper. This pairing of a well-recognized metric with real draw back publicity is a big a part of why the mannequin scaled to tens of thousands and thousands of {dollars} in income inside its first yr.
Precept Three: Share the Threat
Consequence primarily based fashions solely really feel truthful to prospects when the seller accepts among the draw back if outcomes fall brief. This could take the type of a decrease base charge with upside tied to efficiency, service credit for missed targets, or an specific proper to exit if Outcomes should not met inside an agreed interval. A vendor unwilling to share any threat is, in follow, nonetheless promoting entry dressed up in Consequence language.
Precept 4: Make Attribution Express
Each Consequence-based association ought to specify, in writing, how the outcome shall be attributed to the seller’s system versus different elements already at work within the buyer’s enterprise. Ambiguity right here is the most typical supply of disputes and the quickest method to erode belief as soon as a contract is operating.
Precept 5: Deal with the Consequence as a Shifting Goal
Markets, buyer companies, and AI functionality all shift over time. An Outcomes contract signed in the present day ought to embrace a mechanism for revisiting the definition of success, whether or not that may be a quarterly assessment or a renegotiation clause tied to materials modifications within the buyer’s enterprise.
Go-to-Market Practices for Consequence Promoting
Advertising
Advertising for an Outcomes enterprise seems to be completely different from advertising for a software program product in a number of concrete methods.
- Lead with the metric the client already stories to their very own management, not with product options or mannequin functionality.
- Change product demonstrations with proof: audited case research, earlier than and after figures, and third-party validation the place attainable.
- Shift class language away from software program towards managed Outcomes or efficiency partnership, for the reason that true aggressive set is commonly inner headcount or a consulting engagement fairly than one other software program vendor.
- Be conservative in public claims. Overstated Consequence claims are extra damaging to an Outcomes model than they might be to a standard software program model, as a result of the whole pitch rests on credibility of outcomes.
Gross sales
Consequence primarily based gross sales cycles are inclined to run longer on the entrance finish as a result of the client, fairly fairly, needs to grasp how outcomes shall be measured and what occurs if they aren’t achieved. This upfront value is repaid later by materially larger retention, since a buyer who’s renewing primarily based on demonstrated outcomes has a a lot stronger motive to remain than one renewing out of behavior or switching value.
- Construct a joint measurement plan with the shopper in the course of the gross sales course of, not after signature.
- Use a pilot or proof-of-value interval with a small, well-defined workflow and Consequence earlier than committing to a full Consequence primarily based contract.
- Contain finance and operations stakeholders early, since they’re usually those who should validate attribution methodology.
- Compensate gross sales on realized Outcomes for at the very least a portion of variable pay, so incentives don’t encourage overselling functionality the product can not again up.
Contracting
Consequence primarily based contracts want extra precision than conventional license agreements. At minimal they need to specify:
- A written definition of the AI Consequence and the precise system used to calculate it.
- The info sources and AI systems-of-record that shall be used to measure that Consequence.
- The attribution methodology, together with how the impact of the seller’s system shall be remoted from different variables.
- The treatment construction if targets are missed, comparable to charge rebates, service credit, or termination rights.
- A cadence for reviewing and, if vital, renegotiating the Consequence definition.
Case Research: Decagon
Decagon, a conversational AI platform serving prospects with advanced, customized multi step workflows, intentionally selected to not value strictly per Consequence. Charging on a slim per decision foundation created margin threat given how a lot every buyer’s workflow configuration various, and firm management has been candid that parsing the definition of an end result case by case invitations precisely the form of dispute an Outcomes technique is meant to keep away from. As a substitute, pricing is anchored to dialog quantity whereas success remains to be communicated to prospects in Consequence phrases. The lesson for contracting is {that a} defensible, low dispute unit of measure is typically price greater than a purer Consequence metric that can’t be utilized constantly throughout a various buyer base.
The Measurement Layer
The measurement layer is the operational and technical infrastructure that makes an Outcomes technique credible. It’s the a part of the enterprise that solutions three questions on an ongoing foundation: what occurred, how a lot of it may be attributed to the system, and the way assured are we in that attribution. All the things else in an AI Outcomes technique, pricing, contracting, retention, finally will depend on the power of this layer.
Core Parts
| Element | Function | Typical Parts |
| Instrumentation | Seize the uncooked occasions wanted to watch the end result because it occurs | Occasion logging, API telemetry, integration with the shopper’s techniques of document |
| Baseline and Management | Set up what would have occurred with out the system | Pre-deployment baseline interval, holdout teams, matched historic cohorts |
| Attribution Mannequin | Isolate the system’s contribution from different elements | Statistical fashions, incrementality testing, causal inference strategies |
| Verification | Verify the numbers are correct and never manipulated | Third social gathering audit rights, reconciliation with buyer knowledge, anomaly detection |
| Reporting | Current ends in phrases the shopper already trusts | Shared dashboards, agreed reporting cadence, joint log out on figures |
Baselines and Counterfactuals
The toughest and most vital a part of the measurement layer is establishing a reputable counterfactual: what would have occurred if the system had not been used. With out this, any noticed enchancment is simply correlated with the system’s presence, not confirmed to be brought on by it. Frequent approaches embrace operating a baseline interval earlier than deployment, holding out a management group that continues utilizing the prior course of, or constructing a matched comparability from historic knowledge with related circumstances. The best alternative will depend on the shopper’s tolerance for a managed rollout and the scale of the inhabitants being measured.
Attribution Self-discipline
Attribution needs to be handled as a self-discipline with its personal requirements, not an afterthought layered on prime of a dashboard. This usually means separating correlation from causation explicitly in reporting, disclosing confidence intervals fairly than single level estimates, and being keen to report a smaller, defensible quantity fairly than a bigger, contestable one. Distributors that constantly under-claim and overdeliver on Outcomes construct way more sturdy pricing energy than those who do the reverse.
Case Research: Zendesk Automated Decision
Zendesk’s automated decision pricing infers a profitable Consequence from alerts comparable to a buyer accepting a assist heart article, however falls again to a easy rule when no specific sign is accessible: a problem is handled as resolved as soon as there was no additional communication for seventy-two consecutive hours. The hole this exposes is a textbook measurement layer failure. Silence is just not proof of satisfaction. A buyer who quietly provides up and takes their enterprise elsewhere passes the identical automated examine as a buyer who’s genuinely happy, and each are billed identically as a profitable Consequence. The case is a helpful warning {that a} metric will be exactly outlined and constantly measured whereas nonetheless being a weak proxy for the outcome the shopper really cares about.
Governance and Belief
As a result of Consequence primarily based income will depend on figures the seller itself typically calculates, governance issues as a lot as expertise. Sensible mechanisms embrace giving prospects learn entry to the underlying knowledge fairly than solely the completed report, providing third social gathering audit rights for top worth contracts, and publishing a transparent methodology doc that doesn’t change with out discover. Belief within the measurement layer is the precise product being bought in an Outcomes enterprise, much more than the AI system that produces the outcome.
Information and Privateness Concerns
Measurement typically requires entry to delicate operational knowledge, which introduces its personal obligations. A well-built measurement layer minimizes the info it collects to what’s strictly vital for attribution, applies applicable entry controls, and is specific with prospects about what’s retained, for a way lengthy, and for what goal. This isn’t solely a compliance matter. Prospects who belief how their knowledge is dealt with are way more keen to share the info wanted to show Outcomes within the first place.
Organizational Implications
Shifting to an Outcomes mannequin modifications incentives and obligations throughout the enterprise, not solely in gross sales and advertising.
- Buyer success shifts from a retention perform to a income perform, for the reason that staff liable for proving and increasing Outcomes is immediately related to renewal and enlargement income.
- Product and engineering must construct measurement and reporting as top quality options fairly than inner analytics, since prospects will depend on them immediately.
- Finance wants new income recognition practices for contracts the place charges fluctuate with efficiency fairly than utilization.
- Authorized and contracting want standardized language for attribution methodology and cures, so every deal doesn’t require constructing this from scratch.
Suggestions for Working an Consequence Centered Enterprise
The next suggestions summarize methods to put the rules above into follow.
- Construct the measurement layer earlier than scaling Consequence primarily based pricing. A powerful measurement functionality utilized to a handful of accounts is price greater than a broad rollout constructed on weak attribution.
- Begin with pilot engagements that use a slim, well-defined Consequence and a transparent baseline, then increase the mannequin as soon as the methodology has been examined with actual prospects.
- Share threat intentionally. Construction a portion of each Consequence primarily based contract so the seller is uncovered to the draw back of missed targets, not solely the upside of outcomes achieved.
- Give prospects visibility into the info and methodology behind the numbers, fairly than presenting solely a completed report. Transparency is what converts a claimed Consequence right into a trusted one.
- Align inner incentives, notably in gross sales and buyer success, with realized Outcomes fairly than bookings or utilization alone.
- Revisit Consequence definitions on a set cadence. What counts as success in a buyer’s enterprise in the present day might not be the suitable measure in eighteen months.
- Report conservatively. A enterprise that under-claims and constantly delivers builds way more pricing energy over time than one which overstates outcomes and should stroll them again.
- Deal with the measurement layer as core infrastructure, funded and staffed with the identical seriousness because the product itself, since it’s what makes each different a part of the Outcomes technique credible.
Conclusion
Consequence primarily based worth creation is a pure subsequent step as AI techniques tackle extra of the particular work prospects used to pay folks or software program to assist them do. The technique is interesting as a result of it aligns value with worth in a approach that seat primarily based or usage-based fashions hardly ever obtain. It’s also significantly tougher to execute, as a result of it will depend on a measurement layer able to proving, not merely asserting, {that a} outcome occurred and that the seller’s system triggered it.
Companies that make investments early in rigorous baselines, sincere attribution, and clear reporting would be the ones capable of cost for Outcomes credibly and retain prospects who renew as a result of they’ve seen proof, not as a result of they signed a protracted contract.
In an AI pushed market, the measurement layer is just not a supporting perform behind the product. It’s the product that makes the whole lot else sellable.