Review of Impact Analytics, AI-Native Supply Chain Software Vendor

By Léon Levinas-Ménard

Last updated: November, 2025

Impact Analytics is a 2015-founded, venture-backed software company that delivers an AI-branded SaaS suite for retail, grocery, CPG and adjacent industries, focused on demand forecasting, merchandise and financial planning, inventory allocation and replenishment, assortment and space planning, lifecycle pricing, promotions, and business intelligence. Its modules—sold under names like ForecastSmart, InventorySmart, PlanSmart, AssortSmart, MondaySmart and various pricing tools—run as cloud services and are deployed with consulting support and implementation partners for mid-market and large retailers. Over several funding rounds led by Argentum and later Sageview Capital and Vistara Growth, Impact Analytics has raised roughly $60m+ to scale globally, with offices and engineering teams split between the US and India and a client list that includes branded retailers such as Calvin Klein, Tommy Hilfiger, Puma, Lovisa, KiK and Tilly’s. The company markets itself as “AI-native” and increasingly as an “Agentic AI” platform, with a Smart Agent Studio orchestration layer on top of its planning and merchandising modules, but public technical details about the underlying forecasting, optimization and agent architectures remain limited; what can be seen points to a modern cloud and MLOps stack (Kubernetes, Spark, BigQuery/Snowflake, MLFlow/Kubeflow, LangChain-style orchestration) implementing a relatively standard mix of time-series forecasting, machine learning and heuristic optimization tailored to retail, rather than a demonstrably unique state-of-the-art engine.

Impact Analytics overview

Impact Analytics (impactanalytics.co) positions itself as an AI-native, cloud-based planning and merchandising suite for retailers, grocers, consumer brands and supply-chain-intensive businesses. Functionally, it bundles multiple SaaS applications: ForecastSmart for demand planning, InventorySmart for allocation and replenishment, PlanSmart and AssortSmart for merchandise and assortment planning, a family of lifecycle pricing tools, and MondaySmart for business intelligence and anomaly detection.¹²³⁴⁵ All of these sit on a shared data and model layer marketed as AI/ML-driven and more recently as “Agentic AI”, with Smart Agent Studio exposed as a hub to define and orchestrate multi-step agents across workflows. Commercially, Impact Analytics is no longer an early-stage startup: after bootstrapped beginnings it secured an $11m Series A led by Argentum in 2021, followed by additional growth financing and a $40m round in 2024 led by Sageview Capital with Vistara Growth, bringing total funding into the ≈$60m range and supporting expansion across North America, Europe and APAC.⁶⁷⁸⁹¹⁰¹¹¹²¹³ The company has named clients such as Calvin Klein, Tommy Hilfiger, Puma, JoAnn and Belk in earlier disclosures, and more recent partnerships with Lovisa, KiK and Tilly’s demonstrate ongoing adoption of its suite for global fashion, discount and specialty retailers.⁷¹⁴¹⁵¹⁶¹⁷ Technically, the most concrete signals come not from marketing copy but from engineering job postings and architect profiles, which show a fairly standard but up-to-date data and MLOps stack: front-ends in React, services in Python/Node, storage in PostgreSQL plus BigQuery/Snowflake, analytics pipelines on Spark, MLFlow and Kubeflow, and containerized deployments on Kubernetes, with optimization and simulation code written in Python and R. Within this envelope, Impact Analytics appears to implement segment-level time-series forecasting and price/promo optimization with a mix of classical models, ML and heuristics; it is clearly more than a CRUD reporting layer, but there is no public evidence that its algorithms outperform other modern approaches or that its much-repeated claims (e.g. “over one million machine learning models”) reflect something uniquely advanced rather than a large-scale, per-SKU model farm.

Impact Analytics vs Lokad

Impact Analytics and Lokad both operate in the broad area of data-driven supply chain and merchandising decisions, but their philosophies and technical architectures diverge sharply. Impact Analytics is essentially a suite vendor: it offers many pre-packaged SaaS applications (ForecastSmart, InventorySmart, PlanSmart, AssortSmart, MondaySmart, pricing modules, etc.) with configurable parameters and implementation projects, designed primarily for retail and merchandising workflows.¹²³⁴¹⁵ Lokad, by contrast, is a programmable platform built around its own domain-specific language, Envision, where each client’s forecasting and optimization logic is written as code and executed on a custom distributed engine; the product is not a catalog of fixed modules but a supply-chain-specific computing environment used to build bespoke predictive optimization apps.¹⁷¹⁸¹⁹²⁰²¹²²

On the forecasting side, Impact Analytics describes ForecastSmart as an AI-native, ML-driven demand forecasting tool and has marketing copy about handling rare events, short lifecycles and style chaining, but public materials stay at a descriptive level; the company does not publish algorithmic details or benchmarks, beyond noting that its models are award-winning and that it trains a very large number of ML models across its portfolio.¹⁴²³²⁰ Lokad, by contrast, documents a probabilistic forecasting engine that computes full demand distributions (not just point forecasts) across SKUs and locations, including probabilistic lead times, and explicitly states that forecasting is organized as large-scale “tournaments” of models with automatic selection of best candidates.¹⁸²⁰²¹ Lokad’s technical docs further describe the use of differentiable programming and competition-level forecasting techniques to tie forecasts directly to downstream cost functions, rather than optimizing forecast error in isolation.²⁰²¹²² In other words, Impact Analytics markets sophisticated forecasting but treats the modeling layer as an internal implementation detail, whereas Lokad’s publicly documented engine is explicitly distribution-centric and tightly coupled to decision optimization.

On optimization, Impact Analytics clearly does more than safety-stock spreadsheets: its InventorySmart and pricing modules are described as optimization engines that use predictive models and business constraints to generate replenishment, allocation and price recommendations, and the chief architect’s profile mentions simulation and optimization logic implemented in Python and R.¹²²³ However, the exact mathematical formulations (e.g. objective functions, constraints, solvers) are opaque, and there is no independent evidence of how aggressively uncertainty is modeled in the optimization step; the emphasis is on AI-powered applications and, more recently, “Agentic AI” agents driving those apps. Lokad, conversely, builds optimization into the core of the platform: its documentation details stochastic optimization approaches that operate over full probabilistic forecasts, with custom algorithms like stochastic discrete descent and latent optimization, and emphasizes economic drivers (holding cost, stock-out penalty, etc.) as first-class inputs to decision models.¹⁸²⁰²² Instead of separate “modules” for inventory, pricing, etc., Lokad uses Envision code to co-optimize multiple decision types under uncertainty, and publishes its high-level techniques as part of its positioning against other enterprise vendors.²⁰²²

In terms of user experience and deployment, Impact Analytics leans toward a more classical enterprise SaaS pattern: customers license specific modules, work with Impact or partners (e.g. enVista) to integrate data and configure business logic, and then planners use web UIs like MondaySmart dashboards or InventorySmart allocation screens to consume recommendations.¹²⁴¹⁴¹⁶¹⁷ Lokad is closer to a “supply chain IDE”: clients (often through Lokad’s own “supply chain scientists”) write Envision scripts that ingest data, compute probabilistic forecasts and output prioritized action lists; the UI is primarily a cockpit on top of this programmable pipeline, not a gallery of siloed apps.¹⁷¹⁸¹⁹²⁰ Where Impact Analytics is now pushing into LLM-orchestrated “agentic” workflows, Lokad’s innovation emphasis—at least publicly—is still on probabilistic modeling, differentiable programming and stochastic optimization rather than LLM agents; the two firms thus embody different interpretations of “AI in supply chain”: Impact focusing on agent UX and AI-branded vertical apps, Lokad on mathematical rigor and code-driven optimization pipelines.¹⁸²⁰²²

Corporate history, funding and acquisitions

Impact Analytics was founded around 2015 by CEO Prashant Agrawal as a retail-focused analytics and planning company, initially targeting the replacement of spreadsheet-based planning with SaaS tools.⁷¹⁴ In February 2021 the company announced an $11m growth financing (functionally a Series A) led by Argentum Capital Partners IV, with additional participation from other investors; both Argentum’s own release and independent tech press confirm this round and describe Impact Analytics at that time as an AI-driven SaaS provider for planning and merchandising with a global client base including Calvin Klein, Tommy Hilfiger, Puma, JoAnn and Belk.⁶⁷¹²²⁴²⁵ Subsequent articles and funding trackers indicate at least one additional round between late 2022 and early 2023, followed by a much larger growth financing event in January 2024.

On 9 January 2024, Business Wire carried a press release stating that Impact Analytics had closed $40m in growth financing, led by Sageview Capital with additional support from long-time partner Vistara Growth; the release frames Impact as a provider of AI-powered planning and merchandising software for retail, grocery, CPG and supply chain.⁸ Sageview Capital’s own announcement and Vistara’s portfolio news reiterate the same round, reinforcing the headline amount and the identity of the lead investors.⁹⁶¹³ Independent coverage in Indian and US tech and finance outlets—such as VCCircle and IndianStartupNews—corroborates the 40m USD figure, notes that Impact Analytics is a retail SaaS startup with engineering operations in Bengaluru, and places the round roughly 15 months after a previous Series B.¹⁰¹¹²³²¹ The total capital raised across all rounds is reported by secondary sources at about $60m–$62m by mid-2025, though the exact breakdown into Series A/B/growth/Series D is not fully disclosed in primary filings.

No credible evidence could be found of Impact Analytics acquiring another company or being itself acquired; all public announcements relate to funding and partnerships rather than M&A. The frequently cited LOI by a similarly named Canadian microcap “Impact Analytics Inc.” to acquire Antenna Transfer appears to belong to a different entity (Credissential) and is not connected to the retail SaaS vendor reviewed here.

Given its founding date, multiple funding rounds, several hundred employees (per press and job postings), and recurring mentions on growth lists such as the Financial Times’ “America’s Fastest-Growing Companies” and Inc 5000, Impact Analytics should be considered a growth-stage, commercially established vendor rather than an early-stage startup.⁷⁸⁹¹⁴¹³

Product and solution portfolio

Supply-chain and merchandising modules

Impact Analytics’ portfolio is organized around a set of branded SaaS modules that share a common data and AI layer.

InventorySmart is marketed as an “AI-native inventory planning software” that automates allocation and replenishment, aligns stock with demand, and optimizes inventory across channels using advanced forecasting models; the product page emphasizes automated store/DC allocation, multi-channel replenishment, and scenario analysis.²
AssortSmart is described as AI-native assortment planning software for optimizing depth and width of assortments by location and channel to improve margins and inventory turns.³
PlanSmart provides AI-native merchandise financial planning, including open-to-buy budgeting, long-range forecast-driven planning and multi-level plan alignment across product hierarchies.²⁶¹⁵

These modules cover much of classical retail planning (financial planning, assortment, item/size planning, inventory allocation) and are often sold together as an end-to-end merchandising and supply chain suite for fashion and specialty retailers. The Lovisa partnership announcement, for example, states that the Australian jewelry retailer will deploy PlanSmart, AssortSmart, InventorySmart, SpaceSmart and MondaySmart as a fully integrated stack to support its global expansion.¹⁴¹⁵ Similarly, KiK (a German textile discounter) and other European retailers are referenced in Impact’s news feed as adopting combinations of PlanSmart, AssortSmart, ItemSmart and InventorySmart, though not all of these releases were independently retrieved within this review’s scope.

Pricing, promotions and experimentation

Impact Analytics also offers pricing and promotion tools under the broader “PriceSmart” umbrella (BaseSmart, PromoSmart, MarkSmart and TradeSmart in various marketing materials), though fewer third-party sources explicitly list all sub-modules. Product descriptions highlight:

Base price optimization based on demand, competition and margin targets.
Promotion planning and uplift estimation, including cannibalization and halo effects.
Markdown optimization across lifecycle phases.

The “ForecastSmart named Demand Forecasting Solution of the Year” award coverage references Impact’s platform as an end-to-end environment for planning, forecasting, merchandising, pricing and promotions, suggesting that pricing capabilities are integrated with the same underlying forecasting and analytics engine rather than being a separate system.²³²⁰ Some Impact blogs (not cited here to avoid over-quoting) describe Bayesian testing and experimentation concepts in the context of promotions and dynamic pricing, which are later folded into the Agentic AI narrative.

Business intelligence and “Agentic AI”

MondaySmart is positioned as an AI-powered business intelligence layer providing a unified KPI hub and diagnostics for retail performance. Impact’s own solution page describes MondaySmart as identifying key pain points, running in-depth analysis of drivers behind performance deviations, and increasingly leveraging a GenAI “agent” for proactive insights and automation.⁴ The G2 product profile adds that MondaySmart uses machine learning to detect deviations in business performance, analyze promotional effectiveness, and surface insights on how to address underperformance.⁵¹⁷

On top of the core modules, Impact Analytics markets a cross-cutting “Agentic AI” layer and a Smart Agent Studio environment (accessed via a separate subdomain) where users can define agents, tools, data connectors and workflows. While Smart Agent Studio’s UI structure (menus for Agents, Tools, Workflows, Data Connectors, UI Deployments, API Keys, Logs, etc.) suggests a modern LLM/agent orchestration platform, public third-party documentation of its internal workings is sparse; most of what is known comes from Impact’s marketing narrative about agentic workflows driving pricing, replenishment and experimentation.

Technology stack and architecture

Core stack and infrastructure

Because Impact Analytics does not publish detailed system architecture diagrams, the most reliable technical information comes from engineering job postings, architect profiles and secondary write-ups.

A StackOverflow profile for the chief product architect describes the stack as:

Front-end: React.
Back-end: Node.js and Python.
Data stores: PostgreSQL and Google BigQuery.
Simulation and optimization: implemented in Python and R.¹

Senior engineering job descriptions add further detail, listing:

Programming languages: Python, Rust, C++, Java, TypeScript.
Data / MLOps stack: Spark, DuckDB, MLFlow, Kubeflow.
Infrastructure: Kubernetes, Terraform, multi-cloud deployment (AWS, GCP, Azure), Snowflake/BigQuery, Prometheus/ELK for monitoring.¹

Cross-referenced with CioCoverage and other profiles, this paints a consistent picture of a fairly typical mid-2020s AI SaaS stack: a microservice architecture with containerized services orchestrated via Kubernetes, a combination of OLTP (PostgreSQL) and cloud data warehouse (BigQuery/Snowflake) layers, and a Spark-centric data engineering environment for large-scale feature engineering and model training.¹²³¹⁵ No evidence of exotic custom infrastructure (e.g. proprietary storage engines or in-house schedulers) surfaced; Impact appears to rely on mainstream open-source and cloud-native components, which is perfectly reasonable for a vendor of its size.

MLOps and agent platform

Job postings and marketing materials indicate that Impact Analytics uses MLFlow and Kubeflow to manage experiments and deployments, fitting the usual pattern of versioned models, pipelines and serving endpoints. References to LangChain (or similar orchestration layers) and “Agent PaaS” suggest that the Smart Agent Studio is built on top of this MLOps layer, exposing configuration and orchestration of LLM-based agents and tools via a no-code/low-code interface. From the outside, Smart Agent Studio looks similar to other contemporary agent frameworks—structuring agents, tools, data connectors and workflows—but the degree to which it goes beyond orchestration (e.g. planning algorithms, safety rails) cannot be assessed from public information.

Overall, the stack can be summarized as:

React + Node/Python services, PostgreSQL + BigQuery/Snowflake storage, Spark + DuckDB analytics, MLFlow/Kubeflow MLOps, Kubernetes orchestration, with optimization logic in Python/R and an LLM/agent layer orchestrated via Smart Agent Studio.

Machine learning, AI and optimization claims

Impact Analytics’ marketing heavily emphasizes AI and, more recently, Agentic AI. Specific technical claims include:

Use of “over one million machine learning models” across their forecasting and planning portfolio, with automated selection of best-fit models by segment; this claim appears in multiple product pages and awards coverage but is never unpacked into concrete definitions of what counts as a “model” or how the selection is performed.⁴²³²⁰
Advanced forecasting models capable of handling rare events, short lifecycle products and cold starts, including techniques like similarity clustering and “style chaining” for fashion products (described mainly in Impact’s own blogs and white papers).
AI-powered BI (MondaySmart) using ML for anomaly detection, promotion effectiveness analysis and, more recently, GenAI for narrative insights.⁴⁵¹⁷
Reinforcement learning and Bayesian testing for dynamic pricing and promotions (in conceptual blog content).

From a skeptical standpoint, the presence of a modern data and MLOps stack plus explicit mention of optimization code in Python/R supports the conclusion that Impact Analytics does deploy real ML and optimization in production, not just rules and reports.¹²⁵¹⁷ The breadth of retail-specific modules and client base suggests that the models are at least robust enough for mainstream use. However, the public evidence falls short of demonstrating that these models are uniquely state-of-the-art:

There are no public benchmarks (e.g., M-competition-style results) comparing ForecastSmart to open-source baselines or competitor platforms on standard datasets.
There are no open technical white papers that detail model architectures, loss functions, feature engineering pipelines, or optimization formulations.
Claims such as “over one million ML models” are unquantified—this could simply reflect a model-per-SKU/per-store approach, which is conceptually standard in retail forecasting at scale.

In short, Impact Analytics clearly operates a genuine ML/optimization platform built on modern infrastructure, but the depth and novelty of its algorithms remain opaque; based on available information, it is safer to classify its modeling as industry-standard AI/ML for retail planning, not as demonstrably ahead of the research frontier.

Deployment, integration and roll-out

Public case studies and partnership press releases provide some insight into how Impact Analytics is deployed.

The Lovisa partnership release states that Lovisa will use a fully integrated suite (PlanSmart, AssortSmart, InventorySmart, SpaceSmart, MondaySmart) to support rapid global store expansion, implying a multi-module implementation including financial planning, assortment, inventory and BI.¹⁴¹⁵ The Tilly’s announcement notes that Tilly’s will implement InventorySmart and MondaySmart to drive inventory optimization and business intelligence across its stores and distribution centers, with explicit goals of improving in-stock performance and reducing excess inventory.¹⁶¹⁷ Both releases frame Impact’s role as providing AI-native SaaS modules, with the retailer and, in some cases, consulting partners (e.g. enVista in other press not cited here) handling process change and integration with ERP, POS and other systems.

Taken together, the materials suggest a rollout pattern similar to other enterprise SaaS:

Scoping and module selection – choose which Smart modules (ForecastSmart, InventorySmart, PlanSmart, etc.) to deploy.
Data integration – connect ERP, POS, e-commerce and external data sources into Impact’s cloud data layer (BigQuery/Snowflake).
Configuration and calibration – configure hierarchies, constraints, planning calendars and business rules; run pilot forecasts and plans in parallel with existing processes.
Production deployment – expose recommendations via UIs (e.g. InventorySmart screen, MondaySmart dashboards) and integrate outputs with downstream systems (export/import or APIs) for order creation, price updates, etc.
Continuous improvement – iterate models, thresholds and agent workflows based on performance and user feedback.

There is no indication that Impact installs on-premise cores; all references point to multi-tenant cloud deployment. The absence of detailed rollout case studies with timelines means it is impossible to quantify typical implementation durations, but given the complexity of retail merchandising, a multi-month project per client is a reasonable inference.

Client base and sectors

Impact Analytics is clearly focused on retail-centric use cases, especially fashion, specialty and discount retail, with some penetration into grocery and CPG.

The 2021 Series A coverage in Technical.ly lists Calvin Klein, Tommy Hilfiger, Puma, JoAnn and Belk as existing clients, indicating early traction with apparel and craft/fabric retailers.⁷¹⁴
The Lovisa partnership (2025) positions Impact as a key partner for a fast-growing global jewelry retailer, deploying a full suite of planning and merchandising tools.¹⁴¹⁵
The Tilly’s partnership (2025) shows adoption in US specialty apparel, focused on inventory optimization and BI (InventorySmart + MondaySmart).¹⁶¹⁷
Other secondary write-ups and Impact’s own marketing material mention additional European retailers such as KiK and some Italian footwear chains, though independent coverage for each was not exhaustively verified within this review.

Geographically, Impact Analytics is described as US-headquartered (initially in Maryland, more recently reported as New York-based) with a major engineering center in Bengaluru, and client footprints spanning North America, Europe and APAC.⁷⁸⁹¹⁰¹¹²³²¹ The sector and client mix, combined with the size of recent funding rounds and named deployments, support classifying Impact Analytics as a commercially established retail SaaS vendor rather than a niche or experimental player.

Technical assessment and state of the art

From a technology-evaluation standpoint, several points can be made:

Infrastructure and MLOps – Impact’s stack (Kubernetes, Spark, cloud data warehouses, MLFlow, Kubeflow) is aligned with contemporary best practice for data-intensive SaaS and supports the scale implied by its retail clients. There is no indication of lagging infra; if anything, the stack is slightly more modern than that of some long-established APS vendors still tied to on-prem Oracle or monolithic Java apps.⁸⁹¹²¹⁵¹³
Modeling – Impact’s use of ML models, including deep learning for forecasting and ML-based anomaly detection in MondaySmart, appears credible based on job postings and product descriptions, but remains qualitatively described. Without benchmarks or algorithmic detail, these should be considered solid, mainstream ML implementations—likely good enough for most retail use cases but not verifiably better than what a well-resourced in-house data science team or other modern vendors could achieve.
Optimization – The explicit mention of simulation and optimization code in Python/R and the nature of the pricing and inventory modules confirm that Impact goes beyond basic safety stock arithmetic. However, the mathematical form of its optimization problems and the way uncertainty is handled are not documented; it is unclear whether, for example, inventory policies are truly optimized against probabilistic forecasts or are heuristics over point forecasts.
Agentic AI – The Smart Agent Studio and Agentic AI branding show that Impact is investing in LLM-orchestrated agents, likely to automate cross-system workflows (e.g. monitoring KPIs, triggering price simulations, creating tasks). This is in line with broader industry trends, but public technical evidence on agent planning algorithms, safety measures and reliability is currently thin; claims about Agentic AI should therefore be treated as directionally credible but unproven in depth.
Transparency and rigor – Compared to Lokad, which publishes detailed documentation about its probabilistic forecasting, Envision DSL and optimization approaches, Impact offers much less technical transparency. This does not imply its technology is weak, but it limits the ability of an external reviewer to validate “state-of-the-art” claims.

Netting this out, Impact Analytics appears to be:

A modern, cloud-native AI/ML platform for retail planning and merchandising that credibly implements machine learning and optimization at scale, but whose inner algorithms are not publicly documented enough to substantiate claims of being uniquely state-of-the-art.

Discrepancies and evidence quality

A few discrepancies and limitations in the public record are worth noting:

Founding year and HQ – Some secondary profiles list Impact as founded in 2012 or earlier, and company HQ has been reported as both Linthicum Heights (Maryland) and New York City; Technical.ly and funding press releases consistently support a 2015 founding and early operations in Maryland, while more recent aggregator entries report a New York base.⁶⁷⁸⁹²³
Funding totals – Primary sources clearly document the $11m 2021 round and the $40m 2024 round; intermediate rounds and cumulative totals (≈$60m–$62m) are derived from aggregators rather than primary filings and thus should be treated as approximate.⁶⁸⁹¹⁰¹¹¹²²⁴¹³
Client claims – Named clients like Calvin Klein, Tommy Hilfiger, Puma, JoAnn, Belk, Lovisa and Tilly’s appear in independent press or funding coverage, providing reasonably strong evidence; other logos shown on Impact’s own site without third-party corroboration are weaker evidence. Anonymous case studies (e.g., unnamed “global luxury lifestyle house”) are entirely self-reported.
Performance metrics – Reported results such as reductions in lost sales, excess inventory and improved in-stock performance are self-published in case studies and press releases and are not independently audited; they should be treated as indicative but not verified.
Technical depth – The absence of peer-reviewed papers, open technical whitepapers, or open-source core components makes it impossible to fully assess the novelty or robustness of Impact’s models and optimization algorithms.

Overall, the evidence base is typical of a commercial SaaS vendor of Impact’s size: solid for corporate existence, funding and client adoption; reasonably clear for functional scope; thin for deep technical evaluation.

Conclusion

Impact Analytics is a growth-stage, venture-funded SaaS vendor focused on AI-branded retail planning, merchandising and supply chain optimization. It offers a broad suite of cloud modules—ForecastSmart, InventorySmart, PlanSmart, AssortSmart, pricing tools and MondaySmart—implemented on a modern data and MLOps stack and deployed to mid-market and large retailers across multiple regions. Funding rounds led by Argentum and Sageview/Vistara, along with named clients such as Calvin Klein, Tommy Hilfiger, Lovisa, KiK and Tilly’s, confirm that the company is commercially established and operating at meaningful scale. Technically, Impact clearly runs real machine learning and optimization in production and is now layering an “Agentic AI” orchestration platform on top, but it does not expose enough detail for an external reviewer to verify that its algorithms are materially ahead of other modern approaches; the safest interpretation is that Impact delivers competent, industry-standard AI/ML for retail planning packaged in a large suite of vertical applications. Compared with Lokad, Impact’s approach is more module-centric and UX/agent-oriented, whereas Lokad’s is more code-centric and mathematically explicit, with documented probabilistic forecasting and stochastic optimization embedded in a DSL. For buyers, this means Impact Analytics should be evaluated primarily on fit-to-process, data integration, UI and change-management capabilities, with AI claims probed through detailed technical due diligence and empirical testing rather than accepted at face value.