Review of SKU Science, Demand Forecasting Software Vendor

Last updated: April, 2026

Go back to Market Research

SKU Science (supply chain score 3.8/10) is a focused demand-forecasting and forecast-governance SaaS vendor rather than a broad supply chain optimization platform. Public evidence supports a real cloud product with baseline statistical forecasting, multi-level manual adjustments, KPI tracking, ABC or XYZ prioritization, budget uploads, external-forecast comparison, and light collaboration features. Public evidence does not support stronger claims of end-to-end decision optimization, unusually deep machine learning, or a broad autonomous planning engine; the visible center of gravity remains forecast generation, forecast review, and S&OP-style business monitoring.

SKU Science overview

Supply chain score

• Supply chain depth: 3.8/10
• Decision and optimization substance: 3.2/10
• Product and architecture integrity: 4.2/10
• Technical transparency: 4.0/10
• Vendor seriousness: 3.8/10
• Overall score: 3.8/10 (provisional, simple average)

SKU Science is best understood as a lightweight demand-planning application designed to replace spreadsheets for forecasting cycles and forecast-performance review. Its strongest public substance lies in making baseline forecasts quickly, letting users override them at multiple levels, and measuring what those overrides add or destroy. That is useful and commercially coherent, but much narrower than the language of “AI forecasting” can imply.

SKU Science vs Lokad

SKU Science and Lokad both touch forecasting, but they operate at different levels of ambition and software posture.

SKU Science publicly sells a standardized SaaS product with prebuilt demand forecasting, KPI dashboards, collaborative edits, and external-forecast comparison. The user is expected to upload data, review the baseline, adjust forecasts, compare KPIs, and export the resulting numbers back into surrounding business processes. That is the shape of a packaged forecasting layer, not of a general decision-optimization platform.

Lokad is much narrower in module count and much broader in computational ambition. The relevant distinction is not merely that one vendor uses more supply chain jargon than the other. It is that SKU Science’s visible product is centered on forecast review and planner governance, whereas Lokad’s public software posture is centered on computing decisions under uncertainty. On the public record, SKU Science improves the forecasting layer around S&OP; it does not publicly present itself as owning replenishment, allocation, or inventory decisions in the stronger sense.

This difference matters because SKU Science’s strongest evidence is operational convenience and governance around forecasts. Its weakest area is the downstream decision logic that would turn those forecasts into an explicit supply chain action model.

Corporate history, ownership, funding, and M&A trail

SKU Science looks like a small, founder-led French software vendor with a clear topical focus rather than a large planning-suite publisher. The company page identifies three founders: Stephane Leclercq, Thomas Robert, and Nicolas Vandeput, with backgrounds spanning software startups, B2B product work, and supply chain data science. The story presented is one of productizing demand-planning expertise into an affordable cloud service. (4, 14)

The registry footprint is consistent with that story. French company records show SKU SCIENCE as a French SAS created in July 2018, with software publishing as its declared activity and a small headcount. Those sources are not product evidence, but they are important because they anchor the firm as a real operating software entity rather than a loose consultancy brand. (14, 15, 13)

There is little public evidence of large outside funding or acquisition activity. That absence matters for classification: SKU Science reads as a commercially active niche vendor with modest scale, not as a heavily financed consolidation play or a major enterprise-software platform.

Product perimeter: what the vendor actually sells

The current product perimeter is narrower and cleaner than the older page suggested. SKU Science sells baseline demand forecasting, multi-level forecast editing, KPI analysis, dashboards, budget comparison, external-forecast ingestion, product-lifecycle handling, and some multi-user collaboration. The pricing page also makes the product segmentation explicit: simpler plans expose automatic statistical forecasts and aggregation, while higher plans add advanced KPIs, lifecycle automation, promotion management, unlimited scale, and API access. (1, 2, 3)

The help center adds important detail that the marketing pages compress away. The product supports uploading sales history, optionally by item and location, updating new periods cycle by cycle, uploading budgets, uploading external forecasts, switching KPI aggregation methods, and editing forecasts across hierarchy levels with redistribution logic. That evidence turns the product from a brochure into a recognizable operational workflow tool. (16, 17, 18, 19, 20, 21, 22, 23, 24)

The boundary of the offer is also fairly clear. SKU Science is not publicly selling replenishment optimization, network design, supplier planning, production scheduling, or transportation planning. Even where industry pages mention lead times, safety, and inventory effects, the mechanism described is still better demand forecasts and better forecast governance rather than a broader supply chain decision engine. (9, 10, 11)

Technical transparency

Technical transparency is stronger than one might expect from a small forecasting SaaS vendor. SKU Science publishes a substantial help center with concrete usage and configuration details, including how demand data is uploaded, how forecasts are edited, how KPI formulas are interpreted, how external forecasts are compared, and how product lifecycle or multi-user features work. Many larger vendors provide less operational detail in public. (16, 17, 19, 20, 21, 22, 27, 28)

The limit is that this transparency is overwhelmingly about workflow and administration, not about the underlying forecasting engine. The site repeats the “644 statistical combinations” claim and says the engine tests multiple models across time periods, but it does not enumerate the model families, the selection logic, the probabilistic assumptions, or the handling of edge cases such as extreme intermittency and structural breaks. That keeps transparency moderate rather than strong. (1, 2)

Security disclosure is also mixed. The site gives concrete claims about AWS hosting in the EU, encryption at rest, bcrypt password storage, HTTPS, optional 2FA, and database replication. It also prominently lists SOC 2 Type II and ISO 27001 style badges. Still, the public materials do not expose a fuller third-party security dossier or certification references in the reviewable pages, so the evidence remains materially self-attested. (5, 13)

Product and architecture integrity

The product architecture looks coherent in a narrow and practical sense. SKU Science takes in history, produces a baseline forecast, lets users revise it, then measures resulting performance through dashboards and KPIs. The budgets, external forecasts, lifecycle controls, and classifications all fit that central workflow rather than pulling the product in unrelated directions. (2, 3, 16, 19, 20, 22)

System boundaries are reasonably clear. SKU Science is not presented as a transactional system of record and not as a full APS suite. The product is positioned as an overlay that imports data and sends information back out to other tools, which is consistent with the public export and API language. That boundary clarity is a real strength, especially for a smaller vendor. (2, 3, 13)

The main architectural weakness is limited programmatic depth and likely reliance on a fairly bounded product model. Even the more advanced features still read as configurable workflow extensions inside a pre-shaped application rather than as a flexible modeling environment. That is not a defect in itself; it just caps the architectural score in comparison with more programmable systems.

Supply chain depth

SKU Science is meaningfully supply-chain-relevant, but narrowly so. The public material speaks directly to demand planners, S&OP managers, operations managers, and industry verticals where forecast quality matters operationally. The life sciences and industrial-manufacturing pages, in particular, show awareness of lead times, spoilage, service, and production timing. (1, 9, 11)

The positive case is that the product clearly addresses real operational forecasting pain rather than generic analytics. ABC and XYZ prioritization, lag-aware KPI analysis, external-forecast comparison, and aggregate-level forecasting all map to genuine planning work. That makes SKU Science a real peer for the forecasting slice of supply chain software. (20, 21, 25, 26, 27)

The limit is that public supply chain depth stops near forecasting and business review. The product does not visibly own downstream inventory policies, purchasing choices, or execution decisions. The supply chain worldview is competent and useful, but it remains centered on better forecasts and cleaner governance rather than on a broader theory of supply chain economics and optimization.

Decision and optimization substance

This is where the public evidence narrows sharply. SKU Science clearly has more substance than a dashboard shell. It computes baselines automatically, compares user and external forecasts, tracks value added, supports aggregate-level forecast generation, and exposes product-lifecycle and classification features that help focus planner effort. (2, 22, 25, 26, 29)

What the public record does not support is a strong claim of distinctive optimization science. The central engine claim remains the selection of the best forecast among “644 statistical combinations,” but the actual model set, training or selection methodology, and probabilistic depth remain opaque. Even the “advanced AI forecasting” language on pricing is not backed by public algorithmic detail on the reviewed pages. (1, 2, 3)

So the fairest reading is modest but real. SKU Science has genuine forecasting software and a usable forecast-governance layer. It does not publicly read like a state-of-the-art decision engine or a deeply evidenced ML platform.

Vendor seriousness

SKU Science looks like a serious niche product company, even if a small one. The founders’ background, the current pricing structure, the maintained help center, the industry pages, and the named customer stories all support the view that the software is active, supported, and used in practice. This is a stronger seriousness signal than a one-page AI wrapper. (3, 4, 6, 7, 8, 9, 11)

The score is capped because the public rhetoric still leans on performance claims and badges more than on deep technical disclosure. There is also a mild tension between the product’s “ultra quick start” simplicity and the stronger enterprise-grade security and AI language. The product is credible, but the public case is still built more on practicality and founder expertise than on highly falsifiable technical proof. (1, 5, 13)

Supply chain score

The score below is provisional and uses a simple average across the five dimensions.

Supply chain depth: 3.8/10

Sub-scores:

• Economic framing: SKU Science does acknowledge forecast error in monetary terms and pushes users to focus on what matters through ABC or XYZ classes and business review. That is better than treating forecast quality as a purely abstract metric. The economic framing is still downstream and limited, because the product does not publicly tie forecasts to a broader explicit cost-optimization doctrine. 4/10
• Decision end-state: The product is clearly meant to influence business action through forecasts, KPIs, and reviews, and not merely to archive historical data. The end-state is still a human planner adjusting numbers and discussing dashboards rather than software directly producing operational decisions. That supports a moderate score. 4/10
• Conceptual sharpness on supply chain: SKU Science has a clear view of what demand-planning teams need from a lightweight system: fast setup, hierarchy-aware edits, KPI review, and classification-driven focus. That is more precise than generic planning language. It remains narrow and conventional rather than intellectually distinctive across supply chain as a whole, which keeps the score in the lower middle. 4/10
• Freedom from obsolete doctrinal centerpieces: The product is explicitly designed to move companies away from spreadsheet-heavy forecasting and manual report consolidation. That is a real improvement over older demand-planning practices. At the same time, the public doctrine still revolves around consensus forecasts and S&OP-style review rather than a deeper break with traditional planning assumptions. 4/10
• Robustness against KPI theater: SKU Science has real features for measuring value added, bias, and lag-specific forecast performance, which should help reduce superficial KPI management. The public material says much less about avoiding gaming, political overrides, or local optimization around dashboard targets. That gap keeps the score moderate rather than strong. 3/10

Dimension score: Arithmetic average of the five sub-scores above = 3.8/10.

SKU Science is a real supply-chain forecasting tool with meaningful planner utility. The score is capped because its supply chain depth remains concentrated in the forecasting layer rather than in a broader operational decision model. (2, 20, 21, 25, 26)

Decision and optimization substance: 3.2/10

Sub-scores:

• Probabilistic modeling depth: The public material explains model selection, trend and seasonality detection, and hierarchy-aware forecasting. It does not expose probabilistic forecasts, uncertainty distributions, or a clearly articulated stochastic decision framework. That absence keeps the score low. 3/10
• Distinctive optimization or ML substance: The “644 combinations” claim suggests that there is a real bounded forecasting engine inside the product rather than a fake placeholder. What remains missing is public evidence that the engine is technically distinctive beyond standard statistical-model selection, or that the newer AI language materially changes the computational substance. That supports a modest score. 3/10
• Real-world constraint handling: The product does incorporate lead-time-oriented KPI thinking, external-forecast comparison, lifecycle handling, and category prioritization, which are practical demand-planning constraints. Those are useful operational features. They do not amount to a richer optimization model of the supply chain, so the score lands in the lower middle. 3/10
• Decision production versus decision support: SKU Science is a decision-support tool in the literal sense: it helps planners create and review forecasts. It does not publicly claim or show that it computes replenishment, inventory, or scheduling decisions directly. That warrants a moderate but clearly limited score. 4/10
• Resilience under real operational complexity: Named stories and industry pages suggest the product can be used across multiple sectors and forecasting contexts. The product also seems deliberately standardized and fast to deploy, which is a strength. Without deeper public evidence on scale, complexity handling, or unusual forecasting edge cases, the score stays cautious. 3/10

Dimension score: Arithmetic average of the five sub-scores above = 3.2/10.

SKU Science has genuine forecasting substance and a practical governance layer around it. The main limitation is that the public evidence stays close to bounded statistical forecasting and planner review, not to deeper optimization science. (1, 2, 3, 22, 29)

Product and architecture integrity: 4.2/10

Sub-scores:

• Architectural coherence: The public product hangs together well: import data, compute baselines, edit forecasts, compare KPIs, and export or review the results. The adjacent features such as budgets, external forecasts, and lifecycle control all reinforce that central loop instead of distracting from it. This supports a solid score. 5/10
• System-boundary clarity: SKU Science is fairly explicit that it is an overlay product feeding on uploaded data and feeding outputs back into other business systems. That clarity is useful and healthier than pretending to replace ERP or execution software. The score remains moderate because the exact integration surfaces are only lightly described on public pages. 4/10
• Security seriousness: The site does publish concrete operational claims around hosting, encryption, 2FA, and password storage, which is more than empty checkbox theater. The public certification posture is still not richly substantiated in the reviewed material, which keeps the score in the middle rather than higher. 4/10
• Software parsimony versus workflow sludge: SKU Science’s narrow scope is a real virtue. It does not appear to accumulate unrelated modules or giant-suite sprawl. Some workflow weight still exists in the upload, review, and manual-adjustment process, but the product remains comparatively lean. 4/10
• Compatibility with programmatic and agent-assisted operations: Enterprise pricing mentions integrations and API access, and the product clearly supports repeated cyclical data exchange. The public surface still looks much more UI-centric and file-centric than code-centric or automation-native, so the score stays moderate. 4/10

Dimension score: Arithmetic average of the five sub-scores above = 4.2/10.

SKU Science benefits from doing one class of thing reasonably clearly. The product looks coherent and operationally plausible, even if its public architecture is simpler and less programmable than more technical platforms. (2, 3, 5, 13, 16, 24)

Technical transparency: 4.0/10

Sub-scores:

• Public technical documentation: The help center gives real user-facing operational detail on uploads, forecast edits, KPI formulas, and classification behavior. That is meaningfully better than a sparse marketing site. The score is capped because the actual forecasting engine remains only lightly described. 5/10
• Inspectability without vendor mediation: An outsider can understand a substantial portion of how the product behaves in practice from the public tutorials and feature articles alone. What that outsider still cannot inspect is the internals of model selection, training logic, or advanced AI forecasting. That mixed picture supports a moderate score. 4/10
• Portability and lock-in visibility: The upload-and-export posture makes the broad data-flow shape visible enough to infer relatively bounded lock-in compared with a full suite. At the same time, public material does not say much about migration tooling, schema portability, or extraction of model logic, so the score stays in the middle. 4/10
• Implementation-method transparency: The onboarding and recurring-cycle workflow is public, concrete, and easy to understand. The product’s actual implementation posture therefore feels more transparent than average for a small SaaS vendor. It still remains application-level rather than system-internals-level transparency, which keeps the score moderate. 4/10
• Security-design transparency: SKU Science discloses more than just a badge wall by naming encryption, bcrypt, AWS hosting, replication, and 2FA. The public evidence is still mostly self-attested and not deeply audited in the visible pages, so the score remains moderate. 3/10

Dimension score: Arithmetic average of the five sub-scores above = 4.0/10.

SKU Science is moderately transparent in how the application is used and administered. The missing transparency is about the exact modeling and security substantiation behind the stronger forecasting and certification claims. (5, 16, 17, 20, 21, 22)

Vendor seriousness: 3.8/10

Sub-scores:

• Technical seriousness of public communication: The company communicates a clear and bounded product, not a vaporous transformation story. The help center and pricing model also suggest real attention to operational usability. The score still stops below strong because the public technical detail remains shallow where the forecasting engine itself is concerned. 4/10
• Resistance to buzzword opportunism: SKU Science does use AI-flavored language, especially in pricing and founder branding, but the site remains more grounded than many current planning vendors. The product still largely reads as what it appears to be: a practical forecasting tool. That earns a slightly positive score without reaching a strong one. 4/10
• Conceptual sharpness: The company has a coherent view of lightweight demand-planning needs and positions itself clearly against spreadsheet chaos. That is valuable. The conceptual frame remains conventional S&OP and forecast-governance thinking rather than a sharper redefinition of supply chain software. 4/10
• Incentive and failure-mode awareness: Features like value-added tracking, lag-aware KPIs, and baseline versus user comparisons show some awareness that planner overrides can help or hurt. Public material says much less about organizational misuse, bad incentives, or how to prevent political forecast editing. That keeps the score moderate. 3/10
• Defensibility in an agentic-software world: SKU Science’s moat seems to be ease of use, founder credibility, and a focused application rather than unusually deep technical barriers. That can be commercially viable, but it is not a very hard moat on the public evidence. A moderate score is appropriate. 4/10

Dimension score: Arithmetic average of the five sub-scores above = 3.8/10.

SKU Science looks like a serious small vendor with a real product and a credible niche. The score is capped because the public case still relies more on focused practicality and founder authority than on deeply falsifiable technical substance. (3, 4, 5, 14)

Overall score: 3.8/10

Using a simple average across the five dimension scores, SKU Science lands at 3.8/10. That reflects a credible and practically useful forecasting product with respectable public transparency for its size, but limited public evidence of deeper optimization or supply chain decision science.

Conclusion

Public evidence supports treating SKU Science as a real demand forecasting and forecast-governance SaaS vendor. The product looks useful for companies that want a lighter, faster alternative to spreadsheet-heavy forecasting and that value KPI discipline, hierarchy-aware edits, and rapid onboarding more than broad-suite coverage.

Public evidence does not support treating SKU Science as a supply chain optimization platform in the stronger sense. The visible software helps planners produce and review forecasts; it does not publicly own the harder downstream decision layers of supply chain operations. That narrower classification is the most accurate reading of the current public record.