We’re putting $5.5M behind research and development of smaller, specialized AI models for real-world deployment — Read our manifesto →
    All posts

    [Case Study] From ₹20 Lakh to ₹3 Lakh a Month: Rebuilding Enterprise RAG for Real Estate

    How a large real-estate group replaced a costly managed-cloud RAG stack with domain-specific ingestion, retrieval, and an on-premise SLM, reducing its reported monthly run-rate by roughly 85%.

    Conscious Engines

    Industry: Real estate, property operations, and enterprise services
    Client: Large integrated real-estate group, identity withheld
    Use case: Enterprise retrieval-augmented generation across real-estate documents and internal workflows
    Delivered scope: Replacement of the major ingestion, parsing, retrieval, and model layers of an existing managed-cloud RAG system with domain-aware retrieval and an on-premise small language model
    Disclosure: The client has been anonymized. The architecture and ₹20 lakh to ₹3 lakh monthly run-rate figures come from the Conscious Engines engagement record. The figures are presented as reported run-rates, not as an independently audited comparison of matched production periods.

    1. Outcome at a Glance

    The client's enterprise RAG system already worked. The problem was that its economics did not scale.

    Conscious Engines rebuilt the major layers of the system around the client's real-estate language, document structures, and expected outputs. The replacement combined domain-specific ingestion and retrieval with an on-premise SLM, reducing dependence on usage-priced external model services.

    The engagement record reports a reduction in monthly run-rate from approximately ₹20 lakh to approximately ₹3 lakh.

    Key Outcomes

    Approximately ₹20 lakh

    Previous monthly run-rate

    Reported engagement baseline; detailed cost composition is not public.

    Approximately ₹3 lakh

    New monthly run-rate

    Reported rebuilt run-rate; full cost allocation is not public.

    Approximately ₹17 lakh

    Monthly reduction

    Derived from the two reported run-rates.

    Approximately 85%

    Percentage reduction

    Derived as (20 - 3) / 20.

    Approximately 15%

    New cost as a share of baseline

    Derived from the reported run-rates.

    MeasureReported resultEvidence boundary
    Previous monthly run-rateApproximately ₹20 lakhReported engagement baseline; detailed cost composition is not public
    New monthly run-rateApproximately ₹3 lakhReported rebuilt run-rate; full cost allocation is not public
    Monthly reductionApproximately ₹17 lakhDerived from the two reported run-rates
    Percentage reductionApproximately 85%Derived as (20 - 3) / 20
    New cost as a share of baselineApproximately 15%Derived from the reported run-rates
    Annualized baselineApproximately ₹2.40 croreMathematical annualization, not a claim of 12 months of realized spend
    Annualized new run-rateApproximately ₹36 lakhMathematical annualization, not a claim of 12 months of realized spend
    Annualized differenceApproximately ₹2.04 croreScenario value if both monthly figures remain stable for 12 months
    Deployment modelOn-premise SLM with supporting enterprise infrastructureDelivered architecture
    Retrieval designDomain-specific ingestion, parsing, and retrievalDelivered architecture

    The reduction is economically significant. The previous run-rate was approximately 6.7 times the new reported run-rate. But cost alone does not establish success. A valid comparison must show that the rebuilt system processed a comparable workload and maintained an agreed level of answer quality, latency, availability, security, and human-review effort.

    This distinction matters because RAG cost can be made artificially low by answering fewer questions, retrieving less evidence, using stale indexes, accepting lower quality, or moving work to employees. The credible claim is therefore not merely “85% cheaper.” It is:

    The system's reported monthly run-rate fell by approximately 85% after the core RAG architecture was rebuilt. The public claim is limited to the reported run-rate change because matched traffic, quality, and full-cost data are not disclosed.

    That wording is specific, strong, and defensible.

    2. Why a Working RAG System Became Too Expensive

    Enterprise RAG has several cost centers. The language-model API is visible, but it is not the whole bill.

    total RAG cost = ingestion + parsing + embeddings + storage + retrieval + reranking + generation + orchestration + observability + human review

    In a real-estate organization, documents are numerous, heterogeneous, and frequently versioned. A corpus can include agreements, technical reports, operating procedures, property records, vendor documents, project material, commercial correspondence, policies, presentations, and structured system exports. The exact corpus for this engagement is confidential, but the engineering problem is consistent across large property businesses.

    Generic ingestion creates hidden waste

    Real-estate documents do not behave like clean web pages. They contain repeated headers, schedules, annexures, scanned pages, tables, drawings, footnotes, cross-references, signature pages, and nested commercial definitions.

    A generic parser can create several forms of waste:

    • duplicate chunks from headers, footers, and repeated schedules;
    • fragmented tables with lost row and column relationships;
    • clauses separated from the definitions that control their meaning;
    • page-level chunks that are too broad for precise retrieval;
    • sentence-level chunks that lose document structure;
    • repeated re-embedding when only a small part of a document changed;
    • poor metadata that forces the retriever to search the entire corpus;
    • longer retrieved contexts sent to expensive models.

    Every weak ingestion decision creates downstream cost. Bad chunks increase the number of retrieval candidates. Poor retrieval increases reranking and context size. Oversized context increases generation cost and latency. Weak answers create more retries and human review.

    A general model can be expensive in two ways

    The first cost is direct: input tokens, output tokens, requests, provisioned throughput, and related managed services.

    The second cost is architectural. If the model is expected to compensate for poor parsing, weak metadata, missing business rules, and generic retrieval, every request becomes larger and more complex than necessary.

    For bounded enterprise tasks, the strongest general model may be used because the surrounding system is weak. That is an expensive substitute for better system design.

    Cost should be normalized by useful work

    Monthly cloud spend is a useful finance number, but it does not explain unit economics. A production system should measure:

    • cost per query;
    • cost per successfully completed retrieval;
    • cost per grounded answer;
    • cost per answer accepted without rework;
    • cost per document ingested or refreshed;
    • cost per active user;
    • cost per business workflow completed.

    The most useful denominator is usually cost per accepted answer or cost per completed business task. This prevents the team from declaring victory after reducing tokens while answer quality declines.

    The problem in this case was therefore not simply that a cloud bill was high. The existing architecture coupled enterprise usage to a high variable cost without using enough domain structure to control retrieval and inference.

    3. What Was Rebuilt

    Conscious Engines replaced the expensive path through the system rather than applying a single optimization to the existing stack.

    System at a Glance

    Ingestion

    Prepared real-estate content using domain-aware document handling.

    Parsing

    Preserved more useful document structure and metadata.

    Chunking

    Aligned retrieval units with the way the domain asks questions.

    Retrieval

    Ranked evidence using domain terminology and expected outputs.

    Model

    Shifted appropriate inference to an on-premise SLM.

    Orchestration

    Connected retrieval and generation as one controlled pipeline.

    LayerRebuilt functionCost or quality mechanism
    IngestionPrepared real-estate content using domain-aware document handlingReduced noisy and duplicated content entering the index
    ParsingPreserved more useful document structure and metadataImproved retrieval precision and reduced oversized context
    ChunkingAligned retrieval units with the way the domain asks questionsReduced irrelevant passages and repeated model calls
    RetrievalRanked evidence using domain terminology and expected outputsIncreased the chance that the first retrieval contained the right evidence
    ModelShifted appropriate inference to an on-premise SLMReduced recurring external token and API expenditure
    OrchestrationConnected retrieval and generation as one controlled pipelineMade routing, fallbacks, and monitoring possible
    DeploymentOperated private inference inside the client's environmentIncreased control over data movement, capacity, and runtime economics

    The high-level flow became:

    enterprise documents and records

    -> domain-aware parsing and metadata

    -> structured chunks and embeddings

    -> retrieval and optional reranking

    -> on-premise SLM

    -> grounded response for the authorized user

    Domain-aware ingestion did more work before inference

    The redesign moved intelligence earlier in the pipeline. Instead of sending loosely relevant content to a large model and asking it to resolve ambiguity, ingestion and retrieval prepared a smaller, better evidence set.

    For real-estate content, a domain schema can include fields such as document type, property or project, counterparty, jurisdiction, department, effective date, version, confidentiality level, approval status, and source system. The exact schema used in this engagement is not disclosed. These are representative controls that make a production corpus easier to filter and audit.

    The principle is simple:

    better metadata -> smaller search space -> fewer retrieved tokens -> lower inference cost -> clearer citations

    The SLM handled the bounded language task

    The on-premise SLM did not need to reproduce the full capability of a frontier model. It needed to perform the client's recurring retrieval and response task at the required quality.

    Suitable SLM work can include:

    • classifying the user's question;
    • selecting a retrieval route;
    • identifying properties, dates, parties, document types, and other entities;
    • reformulating a query using domain terminology;
    • extracting structured fields;
    • summarizing retrieved evidence;
    • producing a response in the required format;
    • abstaining or escalating when evidence is insufficient.

    This is the practical reason a smaller model can create a large cost reduction. The system narrows the task before the model sees it.

    On-premise did not mean “free”

    Private inference replaces one cost structure with another. The complete post-migration cost should include:

    • compute hardware or cloud-reserved capacity;
    • depreciation or lease treatment;
    • power, networking, and storage;
    • inference-serving software;
    • monitoring and security;
    • model updates and evaluation;
    • platform engineering and support;
    • idle capacity and peak-capacity headroom;
    • any external-model fallback traffic.

    If the reported ₹3 lakh figure excludes material components that were included in the ₹20 lakh baseline, the comparison must be restated. A finance-approved total-cost model is more persuasive than a narrow API comparison.

    The model was one part of the saving

    It would be misleading to attribute the entire reduction to moving inference on-premise. The engagement scope describes a replacement across ingestion, parsing, retrieval, and model layers. Savings can come from fewer indexed duplicates, fewer retrieved tokens, fewer retries, smaller prompts, lower external-model usage, better caching, and more efficient model serving.

    This is why enterprise AI cost reduction is an architectural problem, not merely a model-pricing exercise.

    4. How Production Economics Should Be Measured

    Before the headline becomes a public claim, the old and new systems should be compared over matched periods or through a controlled replay.

    Normalize the workload

    At minimum, record:

    Normalization variableWhy it matters
    Queries per periodA quieter system naturally costs less
    Active usersAdoption changes infrastructure and support load
    Input and output lengthLonger questions and answers change inference cost
    Documents indexedCorpus size affects parsing, embedding, storage, and retrieval
    Documents refreshedUpdate frequency affects recurring ingestion cost
    Retrieval depthMore candidates and reranking increase compute
    Availability targetRedundancy and peak headroom have real cost
    External fallback rateSome traffic may still use paid external models
    Human-review minutesLow automated cost can hide transferred labor

    A controlled replay is particularly useful. The team can run the same representative query set through both architectures, then compare cost and quality under the same traffic distribution.

    Measure quality by failure mode

    One blended “accuracy” number is inadequate for enterprise RAG. The evaluation should separate:

    • retrieval recall: did the correct evidence enter the candidate set?
    • ranking quality: did the system place the best evidence high enough?
    • citation correctness: does each citation support the associated claim?
    • groundedness: is the answer limited to retrieved evidence?
    • completeness: did the answer include every required element?
    • abstention: did the system refuse when evidence was absent or conflicting?
    • access correctness: did retrieval respect the user's permissions?
    • format compliance: did the response match the workflow's required structure?
    • human acceptance: was the result usable without material correction?

    The model should be evaluated on the client's own difficult questions, not generic benchmarks. A useful test set includes common queries, rare clauses, conflicting versions, scanned documents, tables, ambiguous entity names, missing evidence, and requests the user is not authorized to answer.

    This is the same principle behind treating the enterprise evaluation set as a strategic asset.

    Report a unit-economics scorecard

    The engagement record supports the monthly run-rate figures. The operating measures required to normalize those figures are not public:

    KPIBaselineRebuilt systemChange
    Monthly total cost₹20 lakh₹3 lakh85% lower
    Queries per monthNot publicly disclosedNot publicly disclosedRequired to normalize cost
    Cost per 1,000 queriesNot publicly disclosedNot publicly disclosedUseful infrastructure measure
    Accepted-answer rateNot publicly disclosedNot publicly disclosedRequired quality measure
    Cost per accepted answerNot publicly disclosedNot publicly disclosedPrimary economic KPI
    p50 and p95 latencyNot publicly disclosedNot publicly disclosedRequired service measure
    Citation correctnessNot publicly disclosedNot publicly disclosedRequired RAG quality measure
    External-model fallbackNot publicly disclosedNot publicly disclosedRequired routing measure
    Human-review minutesNot publicly disclosedNot publicly disclosedRequired full-cost measure
    AvailabilityNot publicly disclosedNot publicly disclosedRequired service measure

    A fully normalized result would read: 85% lower monthly cost at comparable production traffic and accepted-answer quality. Until that matched comparison is available, “reported” and “approximately” are essential parts of the claim.

    Extend the platform only after the base economics are verified

    The rebuilt system can support additional real-estate and facilities workflows, but the following are roadmap opportunities, not claims about the delivered scope.

    Potential next workflowKnowledge requiredDecision or outputKPI
    Lease and agreement retrievalApproved agreements, amendments, definitions, datesCited answer and clause comparisonReview time, citation accuracy, missed obligations
    Property operations copilotSOPs, asset records, vendor contracts, work ordersTroubleshooting and next-action guidanceMTTR, first-time fix, escalation rate
    Vendor and procurement intelligenceContracts, invoices, service levels, performance historyException detection and supplier comparisonLeakage, invoice exceptions, SLA compliance
    Project document assistantReports, approvals, drawings metadata, correspondenceVersion-aware project answerSearch time, rework, decision delay
    Energy and sustainability intelligenceMeter data, tariffs, equipment, occupancy, emissions factorsDriver analysis and recommended interventionkWh per square foot, peak demand, CO2e
    Voice-based field reportingEngineer speech, site context, work order, evidenceStructured report and follow-up actionsDocumentation time, completeness, closure time

    Each additional workflow should reuse the platform but own its evaluation set, permission rules, and business KPI. A common RAG layer does not make every use case equally safe.

    5. What Enterprise Buyers Should Take Away

    This case is useful because the original system was not a failed prototype. It was a functioning RAG implementation with an unsustainable cost structure. The improvement came from rebuilding the system around the actual task rather than continuing to pay a general-purpose architecture to absorb every inefficiency.

    There are six practical lessons.

    1. Start with cost per accepted answer. Monthly spend is the symptom. Unit economics show whether the system becomes more efficient as adoption grows.
    2. Fix ingestion before buying more model capacity. Poor parsing and chunking increase every downstream cost.
    3. Use domain structure to reduce model work. Metadata, filters, document hierarchy, and expected-output schemas create cheaper and more reliable prompts.
    4. Give each model a defined job. A private SLM can handle frequent bounded work. Larger external models can remain available for justified exceptions.
    5. Compare full costs. On-premise compute, platform labor, fallbacks, and human review belong in the new baseline.
    6. Prove quality on the enterprise's own difficult cases. Generic benchmarks cannot establish production fitness for proprietary real-estate documents.

    Production evidence standard

    ClaimEvidence required for a normalized comparison
    ₹20 lakh previous monthly run-rateIncluded services, invoices or cost ledger, tax treatment, and representative period
    ₹3 lakh rebuilt monthly run-rateHardware and depreciation, power, hosting, support, engineering, and fallback inclusion
    Approximately 85% reductionMatched traffic and corpus, comparison dates, and finance approval
    Comparable system qualityRetrieval, citation, groundedness, acceptance, latency, and availability results
    On-premise SLMModel and hardware description if publishable, capacity, utilization, and fallback policy
    Business impactUser adoption, time saved, completed workflows, and avoided external work

    The immediate next step is not another feature. It is a matched production scorecard that turns an impressive cost number into a durable business case.

    For a buyer evaluating the same problem, the useful diagnostic is straightforward: export 30 to 90 days of model, retrieval, ingestion, infrastructure, and human-review activity; build a cost-per-outcome baseline; replay a representative evaluation set; and identify which layers create cost without improving acceptance quality.

    Sources and evidence boundaries

    • Conscious Engines engagement record. It is the source for the previous architecture, rebuilt layers, on-premise SLM, and ₹20 lakh to ₹3 lakh monthly run-rate. The client's identity is intentionally withheld.
    • The 85% reduction, ₹17 lakh monthly difference, and annualized figures are mathematical derivations from the internally reported monthly run-rates. They are not independently audited or evidence of 12 completed months at the new cost.
    • Production traffic, answer quality, latency, availability, hardware cost, and human-review comparisons are not publicly disclosed. The headline should therefore be read as a reported run-rate change rather than a normalized independent audit.