# Keentech > Keentech is an operations infrastructure company. We design, build, and operate business systems that streamline operations and drive measurable growth across transportation, hospitality, and e-commerce. Site: https://keentech.io Updated: 2026-05-03 --- ## About Keentech Keentech is an operations infrastructure company based in the Philippines, serving clients across Southeast Asia. We combine technology, systems design, and execution to help businesses scale beyond traditional outsourcing. We design, build, and operate integrated systems across transportation, hospitality, and e-commerce. Three industries where operational complexity is the bottleneck to growth. Most businesses do not have a growth problem. They have an operations problem. Disconnected tools, manual processes, and lack of visibility create inefficiencies that slow growth. We replace fragmented stacks and outsourced manpower with integrated infrastructure, and we run it with the client. ### Differentiators - **Systems plus execution.** Software-only solutions do not move operations. We pair the platform with the people running it for outcomes, not licenses. - **Outcome-driven operations.** Outsourcing manpower fixes a headcount problem, not a process problem. We design for the metric the client is trying to move. - **Integrated infrastructure.** Fragmented tools force humans to bridge gaps. We replace stitched-together stacks with one operating picture. - **Continuous optimization.** Delivery is the start, not the end. Every operating cycle gets measured, reviewed, and tuned for the next. ## Engagement Process Five steps end-to-end. We do not stop at delivery. We operate and improve. ### 01. Diagnose We map the operation end-to-end, find where work breaks, and quantify the cost of the breaks. ### 02. Design We redesign the workflow, define the systems behind it, and align the team on who owns what. ### 03. Build We build or integrate the platforms, ship the SOPs, and instrument what we promised to measure. ### 04. Operate We run alongside your team for the first cycles. Less handoff, more accountability. ### 05. Optimize Once the system is steady, we tune for cost, throughput, and customer experience, on rolling cycles. ## Services Six services. Three core systems (TMS, PMS, OMS) plus three pillars that design, integrate, and grow them. ### TMS: Transportation Management System URL: https://keentech.io/services/tms/ *From dispatch to last-mile, one operating picture.* We build TMS platforms that give logistics teams real-time fleet visibility, route optimization, and dispatch control, so transport stops being a black box and starts being a lever. **What is inside:** - Real-time tracking - Route optimization - Fleet performance - Delivery visibility - Driver coordination - Dispatch console - Geo-fencing - ETA modeling - Cost-per-mile - Hub network **Outcomes:** - Cut late ETAs without hiring more dispatchers - Reduce empty miles via route optimization - Surface fleet performance leadership can act on **Service FAQ:** - *What is a TMS and when does a logistics business need one?*: A Transportation Management System centralizes dispatch, fleet visibility, route planning, and proof of delivery. Most logistics businesses need one once dispatch volume exceeds what spreadsheets and chat groups can track without errors. - *How is your TMS different from off-the-shelf options?*: We build it around your operating model: how dispatchers actually work, how routes are decided, how drivers report. Off-the-shelf TMS forces your team to adapt to the software. ### PMS: Property Management System URL: https://keentech.io/services/pms/ *From booking to checkout, every guest interaction tracked.* We design PMS platforms that hospitality teams actually use, with bookings, channel sync, housekeeping, and guest experience working as one system rather than five disconnected tools. **What is inside:** - Booking management - Guest database - Property dashboard - Automated workflows - Channel manager - Rate engine - Housekeeping - Guest experience **Outcomes:** - Eliminate double-bookings via channel sync - Cut front-desk override workflow by 40 percent or more - Surface guest patterns that drive repeat stays **Service FAQ:** - *What does a custom PMS solve that an off-the-shelf one doesn't?*: It maps to how your front desk, housekeeping, and revenue team actually work, and it integrates with your existing channel manager and accounting stack instead of forcing a swap. - *How long does PMS implementation take?*: A typical phase one rollout (bookings plus housekeeping plus reporting) ships in 8 to 12 weeks. We start with the highest-pain workflow first, not the full feature set. ### OMS: Order Management System URL: https://keentech.io/services/oms/ *One inventory truth across every channel.* We build OMS platforms that give e-commerce operators a single inventory source of truth across marketplaces, websites, and stores, plus the fulfillment workflows to ship faster without breaking promises to customers. **What is inside:** - Multi-channel sync - Inventory mgmt - Fulfillment - Order visibility - SKU master - Returns and RMAs - Warehouse routing - Pick and pack **Outcomes:** - End oversells across Lazada, Shopee, Shopify, and store - Cut fulfillment time per order via warehouse routing - Make returns predictable and auditable **Service FAQ:** - *When does an e-commerce business need an OMS?*: Once you sell on more than one channel and run out of patience with manual inventory updates, or you've had your first oversell incident. Whichever comes first. - *Does your OMS integrate with Lazada and Shopee?*: Yes. We treat marketplace integration as a first-class concern, not an add-on. Local channels matter more than global ones for most PH operators. ### Business Operations and Systems URL: https://keentech.io/services/business-ops/ *Process design, automation, and the metrics that prove it works.* We map your operation end-to-end, find the breaks, and redesign the workflow so every handoff has an owner and a measurable output. Systems plus execution, not just slides. **What is inside:** - Process design - Workflow automation - Systems integration - SOP library - RACI mapping - Ops dashboards - KPI design - Cost-to-serve **Outcomes:** - Cut manual handoffs in the workflows that bleed time - Stand up SOPs that survive turnover - Make ops performance visible to leadership weekly **Service FAQ:** - *How is this different from management consulting?*: We don't leave a deck. We instrument the workflow, automate what we can, and measure what we promised. The deliverable is the running system, not the recommendation. ### Software and Web Development URL: https://keentech.io/services/software-web/ *Custom platforms, APIs, and the architecture under both.* When the off-the-shelf tool doesn't fit your operation, we build the platform that does. Web platforms, admin interfaces, mobile-ready UIs, all on architecture designed to scale with you. **What is inside:** - Custom platforms - API integrations - System architecture - Web platforms - Admin interfaces - Mobile-ready UI - Cloud deployment - Observability **Outcomes:** - Replace fragile spreadsheet-and-Drive workflows with a real platform - Integrate the tools your team already uses - Ship faster than rebuilding from scratch on a SaaS that won't bend **Service FAQ:** - *Do you build mobile apps?*: We build mobile-ready web platforms by default, and native mobile when the operation needs offline mode or device hardware. We pick the simpler option that fits the operation. ### Digital Growth Systems URL: https://keentech.io/services/digital-growth/ *Performance marketing, instrumented as a system.* Customer acquisition, conversion optimization, and retention loops built as systems instead of one-off campaigns. Attribution, experimentation, lifecycle messaging, all wired to outcomes leadership can read. **What is inside:** - Performance marketing - Conversion optimization - Acquisition systems - Funnel analytics - Retention loops - Lifecycle messaging - Attribution - Experimentation **Outcomes:** - Acquisition cost trending down quarter over quarter - Retention loops that compound, not stall - A board-readable view of where growth is actually coming from **Service FAQ:** - *Do you do paid media management?*: Yes, but only as part of a full system. We don't run ads in isolation from analytics, lifecycle, and the operations side that fulfills what the ads sold. ## Solutions We deliver three core systems across our target industries: - **TMS for Transportation.** Real-time visibility, optimized routing, fleet management. We run it with the client, not just for them. - **PMS for Hospitality.** Centralize bookings, guest management, property oversight. Less manual reconciliation, fewer errors, higher occupancy. - **OMS for E-commerce.** Unify orders, inventory, fulfillment across channels. One system of record for every SKU, every channel, every order. ## Team ### Darl (Brean) Abrea, Managing Director and Co-founder Sets the operating model and runs the business end-to-end. Bridges strategy, delivery, and execution. ### Ruel Lago, CTO and Co-founder Translates business needs into systems that scale. Leads platform architecture, integration, and reliability. ### Joel Camposano, CRO and Co-founder Aligns commercial strategy to operational reality. Leads partnerships, growth, and client success. ## Frequently Asked Questions ### What kind of operations does Keentech support? Transportation and logistics, hospitality, and e-commerce, plus the operations layer underneath all three. We focus on businesses past their first product but not yet built for scale. ### How is Keentech different from a software development agency? We don't stop at shipping software. We design the operation around it, train the team running it, and stay accountable to the metrics it was supposed to move. ### Where is Keentech based? We are based in the Philippines and serve clients across Southeast Asia. We understand local channel dynamics (Lazada, Shopee, GCash, local logistics) firsthand. ### Do you build custom platforms or use off-the-shelf tools? Both, depending on what fits the operation. We default to integrating tools your team already uses and only build custom when off-the-shelf forces compromises that hurt the workflow. ### What does an engagement typically look like? A diagnostic phase to map the current operation, a design phase aligning systems and SOPs, a build or integration phase, then a live operating phase where we run alongside your team. ### How long until we see measurable improvement? Most engagements show their first measurable wins in the first 8 to 12 weeks. We design the rollout to ship the highest-pain workflow first, not the most feature-complete one. ### Do you only work with large companies? No. We work with businesses past initial product-market fit who are starting to feel scale-related operational pain. The operation has to be real and growing for the work to compound. ### How do you charge for engagements? A fixed-scope fee for the design and build phases, then a monthly operating retainer for the live phase. We size the retainer to the operation, not by counting seats. ## Articles ### From dispatch to dashboard: rebuilding logistics ops without a six-figure platform URL: https://keentech.io/articles/dispatch-to-dashboard/ Published: 2026-04-15 Author: Ruel Lago (CTO and Co-founder) Category: logistics *What we learned routing real fleets, and the trap of optimizing routes without first fixing visibility.* ## TL;DR Most logistics teams reach for route optimization too early. Before optimization, you need visibility. Before visibility, you need clean dispatch data. We rebuilt a regional logistics provider's operations stack from chat-group dispatch and spreadsheet ETAs to a live console in eight weeks. The lift was not the routing algorithm. It was finally seeing what the fleet was doing. ## The problem nobody admits The dispatch desk we walked into was not lacking software. It had three tools: a fleet GPS provider, a spreadsheet of shipments, and a Viber group with the drivers. None of them talked to each other. When a customer called about a late delivery, the dispatcher had to: 1. Search the spreadsheet for the shipment ID 2. Open the GPS provider's web portal and find the truck 3. Message the driver directly to confirm 4. Reply to the customer Each call took three to seven minutes. The desk handled forty calls on a normal Wednesday. That is two and a half hours of a senior dispatcher's day spent on data archaeology before any operational decision gets made. The team's instinct was to fix the routing. The routing was fine. The visibility was the bottleneck. ## What we shipped first, deliberately Before we touched optimization, we built three things that are unglamorous and load-bearing: **A unified shipment record.** Every shipment got an identifier the dispatcher, the driver, and the customer-care team all referenced the same way. No more "the second Cebu run today." Same ID, every conversation. **A live status feed for each shipment.** Not GPS coordinates, just status: scheduled, dispatched, in transit, delivered, exception. The status changed automatically when the driver reported a state on their app, plus manual override for the dispatcher when the app failed. **A dispatcher console** that put the live shipment list, the GPS positions, and the driver chat in one screen. Not a dashboard for leadership yet, the operating console for the desk that runs the day. That was phase one. Six weeks of work. The team was skeptical because none of it looked like the AI-routing demos they had seen at trade shows. Then they used it for a week. ## The actual lift Two numbers moved in the first month after rollout: The customer-care reply time dropped from average four minutes to under thirty seconds. Most calls were answered with a glance at the console. Late ETAs as reported by customers dropped by about a third, even though we had not changed routing yet. The reason: the dispatcher could now spot a delay developing two hours earlier than before, which gave them time to swap a load, reroute, or proactively warn the customer. Visibility creates time. Time is what lets you intervene. ## Then we did the routing Phase two was the algorithm work, eight more weeks. By then the data was clean enough to actually use. Routing optimization on dirty data is worse than no optimization, because the algorithm picks the route based on bad assumptions and the dispatcher loses trust. We added route suggestions, ETA modeling, and cost-per-mile reporting. The team adopted the suggestions in stages. They overrode the algorithm freely for the first month while we tuned the cost function for their reality. By month three, override rate dropped below ten percent. ## What this is not This is not a story about replacing dispatchers. The dispatcher is still there. They are doing fewer Viber lookups and more decisions about exceptions. The console did not save headcount. It bought capacity for a desk that was choking on data plumbing. It is also not a story about a six-figure SaaS. The whole operating console runs on a stack a senior engineer can understand: a Node API, a React console, the existing GPS provider's webhook feed, a Postgres database. Operationally cheap. Architecturally honest. ## When you actually need a TMS If you are a logistics business doing more than fifty shipments a day and your dispatcher's screen has more than three tabs open at once, you need a TMS. Not a SaaS subscription. A console designed around your operating model. The off-the-shelf options force your team to adapt to them; we build it the other way around because the operation already exists and the system has to fit it. If you are doing fewer than fifty shipments and the dispatcher can hold the whole picture in their head, you do not need a TMS yet. You need a clean shipment record and a single source of truth for status. Spreadsheet plus Viber works at that scale, if you discipline how you use them. The trap is the middle: somewhere between a hundred and three hundred shipments a day, where the chaos becomes systemic but the team has not budgeted for the platform. That is when reaction time degrades, customer complaints rise, and senior people start spending all day on data archaeology instead of operations. ## What to do this quarter Before you talk to a TMS vendor or hire a routing consultant, do this: 1. Time-track your dispatch desk for one week. How much of the day is data lookup versus decisions? 2. Audit your shipment IDs. Does every team use the same one for the same shipment? 3. Ask your dispatcher what they would change first if they could change one thing. The answer is rarely "better routing." It is usually some version of "I want to see what is happening without asking three people." Fix that. Then talk about routing. If you want to talk through what this looks like for your operation, [book a consultation](/contact/). We bring the system, but the work starts with understanding yours. --- ### Why your PMS is the bottleneck of your hospitality margin URL: https://keentech.io/articles/pms-bottleneck/ Published: 2026-03-22 Author: Darl (Brean) Abrea (Managing Director and Co-founder) Category: hospitality *The four manual handoffs between booking and check-out that quietly cost you points of GOP every month.* ## TL;DR Most hospitality margin leaks are not pricing problems. They are workflow problems disguised as pricing problems. Four manual handoffs between booking and check-out quietly drain a property's gross operating profit, and they are invisible until you map the flow end-to-end. We have rebuilt this for boutique hotels, serviced apartments, and resort groups. The lift averages two to four points of GOP within one season. ## Where the leaks actually live When a property's GOP is below comp set, the first instinct is to look at rate strategy. Rate strategy is downstream. The upstream issues are these four handoffs: **Booking to confirmation.** A guest books on a marketplace channel (Agoda, Booking, or your own site). Your PMS gets the booking. Your channel manager updates inventory across the other channels. Your accounting system needs the financial record. If any of these three steps lag or fail, you either oversell, lose a booking, or have to manually reconcile later. **Confirmation to arrival.** Pre-arrival communications, payment authorization, room assignment, special requests, housekeeping pre-stage. Five things that happen across email, the PMS, and an Excel file the front office maintains separately. **Arrival to in-house.** Check-in, key issuance, payment finalization, ID verification, registration card storage. The friction at the front desk is the first impression that compounds for the rest of the stay. **In-house to check-out.** Folio reconciliation, mini-bar charges, late check-out approvals, payment, departure communications, post-stay survey trigger. If any of these handoffs require the front desk to manually copy data between systems, you have a leak. The leaks are small individually. They compound across hundreds of stays a month. ## The most common single bottleneck In every PH hospitality engagement we have run, the same handoff is the worst: **booking to confirmation**. Specifically, the channel manager update. Here is what we see: a marketplace booking comes in. The PMS receives it. The channel manager is supposed to push the inventory update to all other channels within a few minutes. But the channel manager is on a separate vendor's platform with a five-minute polling interval. During those five minutes, the same room is technically still bookable on three other marketplaces. Two or three times a month, someone double-books. The override workflow to fix it costs the front office an hour and the guest experience suffers. The fix is rarely a different channel manager vendor. The fix is collapsing the channel manager into the PMS so the inventory update is event-driven, not polled. Five-minute lag becomes five-second lag. Double-booking incidents drop to near zero. ## What we ship in a typical PMS engagement We do not start with software. We start with mapping the four handoffs above as they actually run, not as the org chart says they run. Front desk shadowing, housekeeping ride-alongs, a week of watching the booking inbox. Then we ship in three phases: **Phase one (weeks 1 to 4).** Unify the booking record. One PMS, one source of truth for inventory, one event-driven channel sync. Most of the value lands here. **Phase two (weeks 5 to 8).** Front desk workflows: check-in console, room assignment automation, housekeeping integration. The check-in time per guest typically drops from six minutes to under two. **Phase three (weeks 9 to 12).** Revenue and reporting layer: occupancy tracking, RevPAR, ADR, channel mix, segment analysis. The data is now clean enough to be worth analyzing. We then operate alongside the team for the next two seasons. Phase three reports look different in week thirteen than in month six because the team's behavior changes once the system is honest. ## The numbers that move In an average engagement we see: - Front desk reconciliation time: down 60 to 80 percent - Double-booking incidents: from two to four per month to under one per quarter - Repeat guest rate: up 4 to 8 percentage points within two seasons (because the guest experience at check-in and check-out is materially better) - GOP: up 2 to 4 points within one season The GOP lift is not from rate strategy. It is from labor cost (less reconciliation), from fewer comped stays (fewer overbooking incidents), and from a slight uptick in occupancy because the channel manager actually keeps inventory accurate. ## What to do this quarter, before you talk to a PMS vendor Before you sign anything, run this audit: **Time the front desk.** How many minutes does an average check-in take? Aim for under three. If you are at five or above, the workflow is the issue, not training. **Count overbooking incidents in the last ninety days.** Anything more than once a month is a channel-sync issue, not a strategy issue. **Ask housekeeping how they get the room assignment list.** If the answer is "the front desk WhatsApps it to us," that is a handoff that wants to be automated. **Pull a week's worth of folio reconciliation rework.** How many folios required manual adjustment? If the rate is above 5 percent, the booking-to-confirmation handoff is leaking. The audit is free. The fix is not, but it scopes itself once the audit is done. If you want to walk through the audit for your property, [book a consultation](/contact/). We will bring the framework. The first conversation is on us. --- ### One inventory truth: collapsing four marketplaces into one OMS URL: https://keentech.io/articles/one-inventory-truth/ Published: 2026-02-18 Author: Joel Camposano (CRO and Co-founder) Category: ecommerce *A practical playbook for centralizing multi-channel operations without breaking peak-season fulfillment.* ## TL;DR Multi-channel sellers in the Philippines run inventory across Lazada, Shopee, TikTok Shop, their own Shopify or web store, and one or two physical stores. Most of them keep five separate inventory states and reconcile them twice a day. The first oversell incident usually triggers the OMS conversation. Here is how we run the consolidation without breaking peak-season operations. ## The shape of the problem A typical mid-sized e-commerce operation we work with has roughly these inventory homes: - Lazada seller center inventory - Shopee seller center inventory - TikTok Shop seller inventory - Shopify store inventory - Warehouse management spreadsheet (the actual physical count) Five places. Four of them update on different schedules. The warehouse spreadsheet is the physical truth, but it lags everything by hours. When a customer buys the last unit of a SKU on Shopee, three things should happen: the item ships, the warehouse count decrements, and the other four channels reflect the new zero. In practice, what happens is the item ships, the warehouse count gets manually updated at end-of-day, and one or two other channels still show stock available for ten to thirty minutes after the actual physical zero. Someone buys the now-unavailable unit. The team has to cancel and apologize. Repeat this pattern five to twenty times a week during peak season. ## Why the obvious fix is risky The obvious fix is to install an inventory sync platform that connects all five sources. There are vendors who do this. We have seen them work and we have seen them fail. The failure mode is always the same: the migration weekend. Migration to a centralized inventory truth means freezing every channel for several hours, reconciling counts, then turning sync back on. If your peak season is November to January, you cannot afford that downtime. If you do it during off-peak, the team is not stressed enough to surface the edge cases. Either way, you find out about the edge cases on a Saturday night two weeks after go-live. ## The pattern we ship instead We do not flip everything to centralized at once. We sequence in three phases. **Phase one: pick a primary source of truth.** Usually the warehouse management system, even if it is currently a spreadsheet. The OMS we deploy reads physical inventory from there and treats it as canonical. Phase one ships in two to three weeks. At the end of phase one, the OMS knows the truth but the channels do not yet defer to it. **Phase two: read-and-warn for one week.** We let the OMS observe channel inventory levels and warn when they drift from the truth, but the OMS does not push corrections yet. The team sees a Slack alert when Lazada and the warehouse disagree by more than two units. They confirm or override. This is the most important week of the engagement. The team learns to trust the OMS by watching it be right repeatedly. The OMS learns the team's edge cases. **Phase three: outbound sync.** Once trust is established, the OMS starts pushing corrections to channels. Lazada, Shopee, TikTok Shop, and Shopify each get a webhook contract. We turn on one channel at a time, a week apart. If something goes wrong on Lazada, Shopee is still on the old workflow and the operation does not stop. By the end of phase three, all four channels and the warehouse are sync'd off one source. Total elapsed time: ten to twelve weeks. No frozen weekend. ## The numbers we typically see After full rollout: - Oversell incidents per week: from twelve to sixteen down to under one - Inventory reconciliation labor: from four hours per day across two staff to under thirty minutes per day for one - Order-to-ship cycle: down by 20 to 40 percent because warehouse pickers stop chasing the canonical SKU list across multiple windows - Returns rate: down by a couple of percentage points because fewer wrong items get shipped (the SKU master is now actually canonical) The biggest qualitative win is the team's confidence. The customer service team stops apologizing for cancellations they did not cause. The warehouse stops getting blamed for inventory that was wrong on the channel, not in the bin. ## What an OMS is not It is not a replacement for your warehouse management. The WMS still owns the physical layout, the picking strategy, the bin assignments. The OMS sits above WMS and handles the orders-and-inventory-across-channels layer. It is also not a replacement for your accounting system. Orders flow from the OMS to accounting nightly, not the other way around. And it is not a magic solution if your warehouse count is dirty. We have walked into engagements where the warehouse spreadsheet was off by 8 to 12 percent against a physical recount. No amount of OMS sophistication fixes that. We do a physical recount in week one, full stop. The OMS goes live with truth, not aspiration. ## Should you build or buy? For most operators we work with, the answer is hybrid: buy the marketplace integrations (these are commodity and well-understood), build the orchestration layer (this is where your operation is unique). Off-the-shelf OMS platforms force a one-size-fits-all warehouse model. If you have a Manila hub plus three regional drop points, plus a third-party fulfilment partner for cold-chain, off-the-shelf bends awkwardly around that. We typically ship a custom orchestration layer with vetted off-the-shelf marketplace connectors underneath. You get integration speed and operational fit at the same time. ## What to do before the first OMS conversation A few cheap audits that will sharpen the conversation: 1. Pick one SKU at random. Within thirty minutes, can you tell me the inventory count from each of your five sources? If not, that is your starting baseline. 2. Pull last month's cancellation log. How many were oversell-driven? That number is your floor for the OMS ROI. 3. Time the warehouse team's morning routine. How much of the first hour is reconciliation versus picking? If the audits surface a real operational pain (not just an aspiration to "have a single dashboard"), an OMS engagement will pay for itself within three to four quarters. If they don't, you do not need an OMS. You need cleaner inventory hygiene first. When you are ready to talk through the sequence for your operation, [book a consultation](/contact/). We bring the playbook; the first conversation is on us. --- ## Contact To start a conversation, visit https://keentech.io/contact/. Engagements typically begin with a diagnostic phase to map the current operation, followed by design, build, and operate phases.