<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Syed Ashhad</title>
    <description>The latest articles on DEV Community by Syed Ashhad (@syed_ashhad_b55b32629d2fa).</description>
    <link>https://dev.to/syed_ashhad_b55b32629d2fa</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F2430261%2Fbaf98f2c-23d9-485c-b857-a2fc314787b2.gif</url>
      <title>DEV Community: Syed Ashhad</title>
      <link>https://dev.to/syed_ashhad_b55b32629d2fa</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/syed_ashhad_b55b32629d2fa"/>
    <language>en</language>
    <item>
      <title>Keeping a ZKTeco fingerprint door in sync with gym billing: on-site connector vs ADMS push</title>
      <dc:creator>Syed Ashhad</dc:creator>
      <pubDate>Mon, 16 Nov 2026 14:54:08 +0000</pubDate>
      <link>https://dev.to/syed_ashhad_b55b32629d2fa/keeping-a-zkteco-fingerprint-door-in-sync-with-gym-billing-on-site-connector-vs-adms-push-1djb</link>
      <guid>https://dev.to/syed_ashhad_b55b32629d2fa/keeping-a-zkteco-fingerprint-door-in-sync-with-gym-billing-on-site-connector-vs-adms-push-1djb</guid>
      <description>&lt;p&gt;Most gyms that use fingerprint entry already own a ZKTeco terminal. The hard part of gym software isn't the scanner. It's keeping the scanner's idea of "who may enter" in step with the billing system's idea of "who has paid".&lt;/p&gt;

&lt;p&gt;This is how we approached it in FitZone Pro, the gym management app I build, and the trade-offs between the two ways a cloud app can talk to these devices.&lt;/p&gt;

&lt;h2&gt;
  
  
  The problem
&lt;/h2&gt;

&lt;p&gt;A ZKTeco terminal keeps its own user list and fingerprint templates on the device, and decides at the door whether to open. The membership and billing data lives in the cloud. If the two drift apart, either an unpaid member walks in, or a paid member is stuck outside. The second one is worse.&lt;/p&gt;

&lt;p&gt;So the rule we work to: &lt;strong&gt;the billing record is the source of truth, and the terminal's user list is a cache of "people allowed in right now".&lt;/strong&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Two ways to reach the device
&lt;/h2&gt;

&lt;h3&gt;
  
  
  1. Pull: a small connector on site
&lt;/h3&gt;

&lt;p&gt;Many ZKTeco terminals speak a binary protocol over the local network (TCP port 4370 on most models). A cloud server usually can't reach that port, because the device sits behind the gym's router. So a small agent runs on a Windows computer inside the gym, talks to the terminal over the LAN, and relays to the cloud.&lt;/p&gt;

&lt;p&gt;The non-negotiable requirement: it starts with Windows and reconnects by itself. Nobody at a gym front desk is going to restart a service.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Works with older terminals that have no cloud mode.&lt;/li&gt;
&lt;li&gt;Depends on a PC being on.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  2. Push: ZKTeco ADMS
&lt;/h3&gt;

&lt;p&gt;Terminals that support ADMS ("push" mode) make HTTP requests to a server address you configure: they upload attendance logs and poll for pending commands such as adding or removing a user. No computer at the gym.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;No on-site software to install or babysit.&lt;/li&gt;
&lt;li&gt;Not every model supports it, and a command you queue is only picked up on the device's next check-in, so you design around that delay.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;FitZone Pro supports both. Each branch has one machine, and that machine feeds that branch's attendance.&lt;/p&gt;

&lt;h2&gt;
  
  
  Keeping access in sync
&lt;/h2&gt;

&lt;p&gt;We use two mechanisms, because either one alone fails.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;On payment.&lt;/strong&gt; When the desk records a payment, the member goes back on the terminal straight away. This is the moment the member is standing at the desk waiting, so latency matters most here.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Hourly reconciliation.&lt;/strong&gt; Every hour, each member's status is re-checked against the terminal. This catches plans that expired overnight and any command that didn't land.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;The reason for the second one: a plan expiring is time-based, not event-based. Nothing "happens" in your system at midnight when a monthly plan runs out, so if you only react to events, expired members keep walking in.&lt;/p&gt;

&lt;h2&gt;
  
  
  When the internet drops
&lt;/h2&gt;

&lt;p&gt;The terminal keeps the last list it was given. So paid members still get in during an outage, scans stored on the device sync once the connection comes back, and the desk can check people in by hand in the meantime.&lt;/p&gt;

&lt;p&gt;That is a deliberate choice: people who were allowed at the last sync stay allowed, rather than locking everyone out because the router is down.&lt;/p&gt;

&lt;h2&gt;
  
  
  What I'd tell anyone building this
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Treat the device list as a cache.&lt;/strong&gt; Never decide who has paid by reading the terminal.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Make commands idempotent.&lt;/strong&gt; "Make sure member X is enabled" survives retries; "toggle member X" doesn't.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Show the reason at the desk.&lt;/strong&gt; When a scan is refused, staff should see "plan expired on 16 Sep" or "$45 due", not just "denied". Otherwise the desk overrides the door and the whole system is pointless.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Test on the real hardware.&lt;/strong&gt; Firmware differs between models, and the documentation won't tell you everything.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;If you want to see how this looks from the gym owner's side, the biometric part is described here: &lt;a href="https://fitzonepro.treevoxa.com/biometric-gym-management-software" rel="noopener noreferrer"&gt;https://fitzonepro.treevoxa.com/biometric-gym-management-software&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Disclosure: I'm the founder of Treevoxa, which builds FitZone Pro.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>saas</category>
      <category>iot</category>
      <category>architecture</category>
      <category>webdev</category>
    </item>
    <item>
      <title>How Much Does a Website Cost in Toronto in 2026? Ranges and Why Quotes Differ</title>
      <dc:creator>Syed Ashhad</dc:creator>
      <pubDate>Mon, 16 Nov 2026 13:39:02 +0000</pubDate>
      <link>https://dev.to/syed_ashhad_b55b32629d2fa/how-much-does-a-website-cost-in-toronto-in-2026-ranges-and-why-quotes-differ-3nkn</link>
      <guid>https://dev.to/syed_ashhad_b55b32629d2fa/how-much-does-a-website-cost-in-toronto-in-2026-ranges-and-why-quotes-differ-3nkn</guid>
      <description>&lt;p&gt;For a professionally built small or mid-sized business website in Toronto, a practical planning range is &lt;strong&gt;$5,000 to $20,000 CAD before HST&lt;/strong&gt;, based on published Toronto and GTA pricing guides. A simple landing page can come in under $4,000. E-commerce usually runs $7,500 to $35,000+, and custom web applications start around $25,000.&lt;/p&gt;

&lt;p&gt;This is the short version. The &lt;a href="https://eskore.ca/blog/how-much-does-a-website-cost-toronto" rel="noopener noreferrer"&gt;full Toronto website cost guide on eskore.ca&lt;/a&gt; covers platforms, monthly running costs and a three-year total.&lt;/p&gt;

&lt;h2&gt;
  
  
  Typical ranges (CAD, before HST)
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Website type&lt;/th&gt;
&lt;th&gt;Best for&lt;/th&gt;
&lt;th&gt;Typical cost&lt;/th&gt;
&lt;th&gt;Timeline&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;DIY builder&lt;/td&gt;
&lt;td&gt;Testing an idea&lt;/td&gt;
&lt;td&gt;$0–$1,500/year&lt;/td&gt;
&lt;td&gt;Days to 2 weeks (your time)&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;One-page marketing site&lt;/td&gt;
&lt;td&gt;Single offer or launch&lt;/td&gt;
&lt;td&gt;$1,500–$4,000&lt;/td&gt;
&lt;td&gt;1–3 weeks&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Small business website&lt;/td&gt;
&lt;td&gt;Credibility, basic enquiries&lt;/td&gt;
&lt;td&gt;$4,000–$9,000&lt;/td&gt;
&lt;td&gt;4–8 weeks&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;SEO/content-led service site&lt;/td&gt;
&lt;td&gt;Lead generation in competitive local markets&lt;/td&gt;
&lt;td&gt;$10,000–$25,000&lt;/td&gt;
&lt;td&gt;8–12 weeks&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;E-commerce website&lt;/td&gt;
&lt;td&gt;Selling products online&lt;/td&gt;
&lt;td&gt;$7,500–$35,000+&lt;/td&gt;
&lt;td&gt;8–20 weeks&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Custom web application&lt;/td&gt;
&lt;td&gt;Portals, booking, dashboards&lt;/td&gt;
&lt;td&gt;$25,000–$100,000+&lt;/td&gt;
&lt;td&gt;12–20+ weeks&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h2&gt;
  
  
  Why quotes disagree so much
&lt;/h2&gt;

&lt;p&gt;Every quote is hours × rate × risk:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;40 hours × $75/hr = $3,000&lt;/li&gt;
&lt;li&gt;80 hours × $100/hr = $8,000&lt;/li&gt;
&lt;li&gt;180 hours × $140/hr = $25,200&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;A $3,000 quote and a $25,000 quote aren't pricing the same website. The useful question is what the extra 140 hours are for.&lt;/p&gt;

&lt;h2&gt;
  
  
  What doesn't show in a screenshot
&lt;/h2&gt;

&lt;p&gt;Two sites can look identical on launch day. The differences show up later:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Discovery and copy guidance&lt;/li&gt;
&lt;li&gt;A CMS structured around how you'll actually edit&lt;/li&gt;
&lt;li&gt;Testing on real phones and browsers&lt;/li&gt;
&lt;li&gt;301 redirects from your old URLs&lt;/li&gt;
&lt;li&gt;Analytics with proper conversion events&lt;/li&gt;
&lt;li&gt;Core Web Vitals and accessibility checks&lt;/li&gt;
&lt;li&gt;Someone answering the phone in month three&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Toronto-specific costs
&lt;/h2&gt;

&lt;p&gt;Budget for HST on top of the quote, and for accessibility: AODA rules apply to larger Ontario organisations and are good practice for everyone. If you sell online in Canada, plan for Canadian tax settings and shipping rules in the store.&lt;/p&gt;

&lt;h2&gt;
  
  
  How to compare quotes
&lt;/h2&gt;

&lt;p&gt;Put every quote against the same list: number of pages, who writes the copy, platform, redirects, analytics, accessibility, hosting, and support after launch. Ask what is excluded, not only what is included.&lt;/p&gt;

&lt;p&gt;For a brand-new business, start small: a template or one-page site, then spend more once you know which pages bring enquiries.&lt;/p&gt;

&lt;p&gt;The full guide, with sample quotes and the sources behind these ranges: &lt;a href="https://eskore.ca/blog/how-much-does-a-website-cost-toronto" rel="noopener noreferrer"&gt;How much does a website cost in Toronto?&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Disclosure: I run &lt;a href="https://eskore.ca" rel="noopener noreferrer"&gt;Eskore&lt;/a&gt;, a web and software studio in Toronto. The ranges come from published Toronto pricing guides cited in the full article.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>webdev</category>
      <category>webdesign</category>
      <category>smallbusiness</category>
      <category>wordpress</category>
    </item>
    <item>
      <title>ERP Software for Small Manufacturers in Canada: Costs and Which Route Fits</title>
      <dc:creator>Syed Ashhad</dc:creator>
      <pubDate>Mon, 16 Nov 2026 13:26:28 +0000</pubDate>
      <link>https://dev.to/syed_ashhad_b55b32629d2fa/erp-software-for-small-manufacturers-in-canada-costs-and-which-route-fits-52fa</link>
      <guid>https://dev.to/syed_ashhad_b55b32629d2fa/erp-software-for-small-manufacturers-in-canada-costs-and-which-route-fits-52fa</guid>
      <description>&lt;p&gt;The best ERP for small manufacturers in Canada, for most shops of 5 to 50 people, is a packaged system such as Microsoft Business Central or MRPeasy. Odoo or ERPNext suit shops that want more flexibility and have a partner to set them up. A custom build makes sense when your process is unusual, or when per-user fees keep climbing as you hire.&lt;/p&gt;

&lt;p&gt;This is the short version of a longer guide. The &lt;a href="https://treevoxa.com/guides/erp-software-small-manufacturers-canada/" rel="noopener noreferrer"&gt;full guide on treevoxa.com&lt;/a&gt; has the sources, the module order and the Canadian tax checklist in detail.&lt;/p&gt;

&lt;h2&gt;
  
  
  When does a small manufacturer need an ERP?
&lt;/h2&gt;

&lt;p&gt;Most SMB manufacturers start looking at 10 to 25 staff. Your shop probably needs one if several of these are true:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Stock-outs on some materials and over-ordering on others, and nobody trusts the stock count.&lt;/li&gt;
&lt;li&gt;Quotes based on gut feel or last year's material prices, not the real cost of the job.&lt;/li&gt;
&lt;li&gt;Bills of materials in spreadsheets, with different people holding different versions.&lt;/li&gt;
&lt;li&gt;QuickBooks can't track batches, work orders or what was consumed on which job.&lt;/li&gt;
&lt;li&gt;A customer asks which lot went where, and the answer takes a day of digging.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;A shop under about 10 people often does fine with QuickBooks plus a stock app. An ERP adds discipline: someone has to enter receipts, issue materials and close production orders.&lt;/p&gt;

&lt;h2&gt;
  
  
  Start with three modules, not twelve
&lt;/h2&gt;

&lt;p&gt;Inventory with lot tracking, bill of materials, and costing fix the most money. Purchasing comes next, then production scheduling, sales and quality. Finance can stay in your accounting package and be integrated later.&lt;/p&gt;

&lt;h2&gt;
  
  
  What a 15-user shop pays over three years (CAD, estimate)
&lt;/h2&gt;

&lt;p&gt;Our planning assumptions, not quotes: 15 users, 36 months, packaged cloud ERP at a blended CAD 130 per user per month, Odoo or ERPNext at CAD 45 with a partner, and a custom build adapted from an existing working ERP.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Cost line&lt;/th&gt;
&lt;th&gt;Packaged cloud ERP&lt;/th&gt;
&lt;th&gt;Odoo or ERPNext&lt;/th&gt;
&lt;th&gt;Custom build&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Licence or subscription&lt;/td&gt;
&lt;td&gt;70,200&lt;/td&gt;
&lt;td&gt;24,300&lt;/td&gt;
&lt;td&gt;0&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Implementation&lt;/td&gt;
&lt;td&gt;45,000&lt;/td&gt;
&lt;td&gt;35,000&lt;/td&gt;
&lt;td&gt;20,000&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Customisation or build&lt;/td&gt;
&lt;td&gt;10,000&lt;/td&gt;
&lt;td&gt;15,000&lt;/td&gt;
&lt;td&gt;60,000&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Hosting and support&lt;/td&gt;
&lt;td&gt;15,000&lt;/td&gt;
&lt;td&gt;18,000&lt;/td&gt;
&lt;td&gt;45,000&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Total over 36 months&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;140,200&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;92,300&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;125,000&lt;/strong&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;ul&gt;
&lt;li&gt;&lt;strong&gt;Odoo or ERPNext is the cheapest route here.&lt;/strong&gt;&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Custom overtakes packaged after about 22 months&lt;/strong&gt;, because it costs about CAD 13,400 a year less to run.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Headcount moves the packaged line most.&lt;/strong&gt; Ten more users add CAD 46,800 to the packaged licence over three years.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Swap in your own numbers before you decide.&lt;/p&gt;

&lt;h2&gt;
  
  
  Which route fits you?
&lt;/h2&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;If this sounds like you&lt;/th&gt;
&lt;th&gt;Route&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Fairly standard process, under about 15 users, want a local reseller in English and French&lt;/td&gt;
&lt;td&gt;Packaged ERP&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Want broad features and lower licence cost, and a partner can configure it&lt;/td&gt;
&lt;td&gt;Odoo or ERPNext&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Unusual process, or quoting, costing or traceability rules that don't fit standard modules&lt;/td&gt;
&lt;td&gt;Custom build&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Expect headcount to grow fast and per-user fees would bite&lt;/td&gt;
&lt;td&gt;Custom, or self-hosted Odoo or ERPNext&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;h2&gt;
  
  
  Canadian questions to ask every vendor
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;Show GST, HST, PST and QST applied correctly on a sample order across two provinces.&lt;/li&gt;
&lt;li&gt;Print an invoice, quote and packing slip in French or English by customer.&lt;/li&gt;
&lt;li&gt;Invoice in USD and show CAD margin on the same job.&lt;/li&gt;
&lt;li&gt;Store tariff codes and country of origin per product.&lt;/li&gt;
&lt;li&gt;Who updates tax tables when rates change, and how fast?&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Ask for a working demo with your data
&lt;/h2&gt;

&lt;p&gt;Before you sign, ask the vendor to show your real BOM with a sub-assembly costed correctly, a purchase order received with a lot number, a production order with scrap, a sales order invoiced with the right tax, and a trace from a finished lot back to supplier lots.&lt;/p&gt;

&lt;p&gt;The full guide, with the module table, implementation timeline and sources: &lt;a href="https://treevoxa.com/guides/erp-software-small-manufacturers-canada/" rel="noopener noreferrer"&gt;ERP software for small manufacturers in Canada&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Disclosure: I run &lt;a href="https://treevoxa.com" rel="noopener noreferrer"&gt;Treevoxa&lt;/a&gt;, which builds custom ERP and makes Batchbook, a manufacturing ERP. The cost table above uses our assumptions, so replace them with yours.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>erp</category>
      <category>manufacturing</category>
      <category>smallbusiness</category>
      <category>productivity</category>
    </item>
    <item>
      <title>How to Add a Mobile App: Connecting Your Existing ERP or Web System</title>
      <dc:creator>Syed Ashhad</dc:creator>
      <pubDate>Sun, 08 Nov 2026 09:00:00 +0000</pubDate>
      <link>https://dev.to/syed_ashhad_b55b32629d2fa/how-to-add-a-mobile-app-connecting-your-existing-erp-or-web-system-102a</link>
      <guid>https://dev.to/syed_ashhad_b55b32629d2fa/how-to-add-a-mobile-app-connecting-your-existing-erp-or-web-system-102a</guid>
      <description>&lt;p&gt;A mobile app can often sit on top of an existing ERP or web system without replacing it. This guide covers the integration routes, what the system must provide and what drives cost.&lt;/p&gt;

&lt;p&gt;Yes, in most cases you can add a mobile app to an existing ERP or web system without replacing it, provided its API covers the jobs your staff need on a phone. Treevoxa builds these apps with &lt;a href="https://treevoxa.com/services/react-native-app-development/" rel="noopener noreferrer"&gt;React Native app development&lt;/a&gt;. One React Native code base serves iOS and Android, with the same API behind both. This guide explains how to add a mobile app to an existing ERP or web system: which integration route fits, what the current system must provide, and what drives cost and timing.&lt;/p&gt;

&lt;p&gt;The phone app and the desktop system read and write through the same backend. A stock adjustment made on a phone and one made at a desk land in the same database and pass the same checks. The questions are the same whether your head office is in Thornhill or Denver.&lt;/p&gt;

&lt;h2&gt;
  
  
  1Can you add a mobile app to an existing ERP or web system?
&lt;/h2&gt;

&lt;p&gt;Usually, yes. The deciding question is not whether the system has an API, but whether that API covers the jobs your people need on a phone: checking stock, posting a receipt, approving a purchase, updating a job or recording a delivery. If it does, an ERP mobile app is mostly new screens over existing endpoints. If it does not, the gap is the first thing to fix.&lt;/p&gt;

&lt;p&gt;The app holds no business rules of its own. It calls the same backend the office uses, so the desk and the phone follow the same rules.&lt;/p&gt;

&lt;h3&gt;
  
  
  Which endpoints and workflows must you map first?
&lt;/h3&gt;

&lt;p&gt;Before anyone draws a screen, audit what you have. Look at the front end, the back end, the APIs, the database, how users log in, and what each role is allowed to do.&lt;/p&gt;

&lt;p&gt;Then list the five to ten tasks staff need on a phone. For each task, find the endpoint that performs it. Every task without an endpoint is a gap, and gaps set both scope and cost.&lt;/p&gt;

&lt;p&gt;Take a hypothetical receiving clerk at a distributor in Raleigh. The clerk needs to scan a purchase order, post the quantities received and flag a short delivery. If the API can read purchase orders but cannot post receipts, the app cannot do that job until the receipt endpoint exists. That endpoint goes into the scope at the start, so the price reflects it.&lt;/p&gt;

&lt;h3&gt;
  
  
  Which system should hold the record?
&lt;/h3&gt;

&lt;p&gt;This is a business decision, and it comes before design. When the phone and the office both change the same record, one system must be the source of truth.&lt;/p&gt;

&lt;p&gt;Ask three questions. Where is a stock count saved? Who can reverse an approval? Which edit wins when a customer note changes in two places?&lt;/p&gt;

&lt;p&gt;In our projects, we usually recommend that the ERP holds stock, orders and approvals, while the app holds drafts until they are posted. Whatever you choose, write the answer down for each record type. Unwritten rules are the ones people argue about later.&lt;/p&gt;

&lt;h2&gt;
  
  
  2What if your system has no usable API?
&lt;/h2&gt;

&lt;p&gt;Older ERPs, in-house web applications and packaged systems with thin APIs are common. There are four routes in, from a documented API to direct database access. Where the source is awkward, middleware or a custom connector sits between the app and the system and does the translating.&lt;/p&gt;

&lt;p&gt;Screen-scraping, where software clicks through the existing screens, can work for a while. It breaks when the page layout changes. Treat it as a last resort.&lt;/p&gt;

&lt;p&gt;A web product that renders its pages on the server, or whose API is tightly coupled to those pages, needs backend work before a mobile app is feasible. You cannot simply point the app at it. The logic has to be pulled out into a service the app can call first. Scope that groundwork before anyone quotes a launch date.&lt;/p&gt;

&lt;p&gt;Take a hypothetical property maintenance company in Austin whose technicians update jobs in a web portal built around page templates. The portal works in a browser, but it has no endpoint a phone could use. The first project is not the app. It is the service layer the app would later call.&lt;/p&gt;

&lt;p&gt;If you own a custom-built system, the cleanest answer is usually to add an API layer to it. For systems built to your requirements, see &lt;a href="https://treevoxa.com/services/erp-software-development/" rel="noopener noreferrer"&gt;ERP software development&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  Which integration route fits your system?
&lt;/h3&gt;

&lt;p&gt;Effort and risk are relative. Your system and your workflows decide where each route sits. The simplest form of API integration is a documented API, and the other routes are fallbacks.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Route&lt;/th&gt;
&lt;th&gt;Effort&lt;/th&gt;
&lt;th&gt;Risk&lt;/th&gt;
&lt;th&gt;Best when&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Documented API&lt;/td&gt;
&lt;td&gt;Low&lt;/td&gt;
&lt;td&gt;Low&lt;/td&gt;
&lt;td&gt;The system has a current API that covers the workflows the app needs&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Middleware or custom connector&lt;/td&gt;
&lt;td&gt;Medium&lt;/td&gt;
&lt;td&gt;Medium&lt;/td&gt;
&lt;td&gt;The API is partial, or two or more systems must stay in step&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;File exchange (scheduled import and export)&lt;/td&gt;
&lt;td&gt;Low&lt;/td&gt;
&lt;td&gt;Medium&lt;/td&gt;
&lt;td&gt;Data can wait minutes or hours, and nobody needs live updates&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Direct database access&lt;/td&gt;
&lt;td&gt;High&lt;/td&gt;
&lt;td&gt;High&lt;/td&gt;
&lt;td&gt;Rarely. Only when no API exists and the vendor's licence terms allow it&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Direct database access deserves the most caution. Stock limits, approval steps and required fields usually live in the application, not in the tables. Write straight to the tables and those checks are skipped. The errors often surface later as wrong stock figures, and nobody can trace which change caused them.&lt;/p&gt;

&lt;h3&gt;
  
  
  Why does a thin API layer often come first?
&lt;/h3&gt;

&lt;p&gt;A thin API layer is a small service between the legacy system and the mobile app. The app talks only to that layer, and the layer talks to the legacy system, however awkward it is. In our projects, this layer usually comes first, before any screen is designed.&lt;/p&gt;

&lt;p&gt;It also lets you replace the old system later. When the ERP is eventually swapped out, you rewire the thin layer, and the mobile app keeps calling the same endpoints. The app does not need rewriting. That is the main reason to build the layer.&lt;/p&gt;

&lt;p&gt;The layer should expose only what the app needs, which keeps the scope small. Treevoxa builds these layers under &lt;a href="https://treevoxa.com/services/api-development/" rel="noopener noreferrer"&gt;API development&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  3Should the mobile app write straight into the system?
&lt;/h2&gt;

&lt;p&gt;Usually not. Writes should go through an API that enforces the business rules: stock cannot fall below a set limit, an approval needs the right role, and a job cannot close without a signature. If the app writes past that API, those checks are skipped too.&lt;/p&gt;

&lt;p&gt;Reads can be more direct, but route them through the API as well, so permissions apply to every request. If customer records are part of the project, &lt;a href="https://treevoxa.com/services/crm-development/" rel="noopener noreferrer"&gt;CRM development&lt;/a&gt; covers how the two systems connect. &lt;a href="https://treevoxa.com/guides/crm-integration-with-erp/" rel="noopener noreferrer"&gt;CRM integration with ERP&lt;/a&gt; sets out the methods and a worked flow.&lt;/p&gt;

&lt;h2&gt;
  
  
  4Should you build a native app, a wrapper or a progressive web app?
&lt;/h2&gt;

&lt;p&gt;There are three realistic options. Pick by what your staff need, not by which one is quickest to start.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Native app built with React Native.&lt;/strong&gt; One code base serves iOS and Android. It has full access to the camera, barcode scanning, push notifications and local storage, so offline use is strong. Build effort is the highest of the three, but one shared code base keeps it well below two separate native builds. It gets a store listing.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Web wrapper.&lt;/strong&gt; The existing web app is packaged in a shell for the app stores. Build effort is low. Device access is limited to what the wrapper exposes, offline use depends on what was cached, and store approval is not guaranteed.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Progressive web app (PWA).&lt;/strong&gt; The app runs in the browser and can be added to the home screen. It reuses the web code, so build effort is low. Device access is narrower, and some browser features on iOS lag behind Android. There is no store listing. Offline use covers cached pages and queued actions.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;For apps built to your brief, see &lt;a href="https://treevoxa.com/services/mobile-app-development/" rel="noopener noreferrer"&gt;mobile app development&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  Why can app stores reject a thin wrapper?
&lt;/h3&gt;

&lt;p&gt;Apple's App Review Guideline 4.2, on minimum functionality, can reject an app that is no more than a repackaged website. The point of the rule is value. The app should do something a browser tab cannot.&lt;/p&gt;

&lt;p&gt;A good wrapper adds features that need a native app: offline use, device features such as the camera or barcode scanning, or push notifications. A wrapper that shows a login screen and nothing else is the kind that gets turned away. Apple's wording changes, so check the current guideline on developer.apple.com before you commit to a store launch.&lt;/p&gt;

&lt;p&gt;Google Play has a separate rule. From 31 August 2026, new apps and app updates must target Android 16 (API level 36) or higher. That date has passed, so any plan that targets an older level needs changing before submission. Check Google's current wording on its &lt;a href="https://support.google.com/googleplay/android-developer/answer/11926878?hl=en-GB" rel="noopener noreferrer"&gt;Play Console help page&lt;/a&gt;.&lt;/p&gt;

&lt;h3&gt;
  
  
  When is a progressive web app enough?
&lt;/h3&gt;

&lt;p&gt;A progressive web app is often the right answer for internal tools. It is the fastest route, and it avoids store review.&lt;/p&gt;

&lt;p&gt;A progressive web app suits:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Internal forms such as timesheets, inspections or order entry&lt;/li&gt;
&lt;li&gt;Shared devices on a warehouse floor, in a shop or on a site&lt;/li&gt;
&lt;li&gt;Staff who need no push notifications and no heavy camera use&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Take a hypothetical contractor in Denver logging site inspections on shared tablets. That is a good PWA candidate.&lt;/p&gt;

&lt;p&gt;Be aware of:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Less control over device features, especially on iOS&lt;/li&gt;
&lt;li&gt;No store listing, so staff open a link or add it to the home screen&lt;/li&gt;
&lt;li&gt;Narrower offline support than a native app&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  5How much will it cost, and how long will it take?
&lt;/h2&gt;

&lt;p&gt;We do not quote a figure in this article. A number without a defined scope is a guess, and a guess is no use to a budget holder. What we can give you is what moves cost and time, and how to get a real number.&lt;/p&gt;

&lt;p&gt;Start with the &lt;a href="https://treevoxa.com/tools/software-cost-calculator/" rel="noopener noreferrer"&gt;software development cost calculator&lt;/a&gt; for a first estimate. For rate context, see &lt;a href="https://treevoxa.com/software-development-rates/" rel="noopener noreferrer"&gt;software development rates&lt;/a&gt;. &lt;a href="https://treevoxa.com/guides/erp-development-cost/" rel="noopener noreferrer"&gt;ERP development cost&lt;/a&gt; explains what a custom ERP really costs.&lt;/p&gt;

&lt;p&gt;Timing follows the same drivers as cost. A single read-only screen over existing endpoints takes far less time than a multi-role app with offline sync and three integrations. Your endpoint audit tells you where in that range you sit.&lt;/p&gt;

&lt;p&gt;Treevoxa agrees a written scope, a fixed price and a launch date before work starts. You know what is being built, what it costs and when it ships before any code is written.&lt;/p&gt;

&lt;h3&gt;
  
  
  What drives the cost up or down?
&lt;/h3&gt;

&lt;p&gt;When you turn a web app into a mobile app, the same six drivers apply. Check these first:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;API coverage.&lt;/strong&gt; How many of the phone tasks already have endpoints?&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;User roles.&lt;/strong&gt; Each role changes what the screens show and what the API allows.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Offline support.&lt;/strong&gt; Sync is among the hardest parts to build and test.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Data volume.&lt;/strong&gt; Large catalogues or long histories need paging and careful caching.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Integrations.&lt;/strong&gt; Each other system the app reads from or writes to adds work.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Design needs.&lt;/strong&gt; A plain internal tool costs less than a branded, customer-facing app.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;React Native helps on the build side. One code base serves both platforms, so you are not paying for two separate native apps.&lt;/p&gt;

&lt;p&gt;If your web system is built in React, see &lt;a href="https://treevoxa.com/services/react-development/" rel="noopener noreferrer"&gt;React development&lt;/a&gt;. Validation rules, API client code and type definitions can often be shared with the React Native app. That removes a second round of the same work.&lt;/p&gt;

&lt;h3&gt;
  
  
  What should a first release cover?
&lt;/h3&gt;

&lt;p&gt;Pick one module or task set for version one. Stock checks, job updates or delivery sign-off are good candidates: they are frequent, clearly defined and easy to measure.&lt;/p&gt;

&lt;p&gt;Hold everything else for later phases. A first release that does three things well tells you more than a broad one that is half-finished.&lt;/p&gt;

&lt;p&gt;Agree four things before work starts: the scope, the fixed price, the launch date and who owns what at handover. Treevoxa's position is that you own the code and the accounts, including the app store and cloud accounts.&lt;/p&gt;

&lt;h2&gt;
  
  
  6Should you build a custom app or use your ERP vendor's?
&lt;/h2&gt;

&lt;p&gt;A custom-built system has no vendor SDK to extend, so the mobile architecture starts from the API. There is no ready-made mobile framework to lean on, which is why the endpoint audit comes first.&lt;/p&gt;

&lt;p&gt;A packaged ERP may come with its own mobile app. That helps, but the vendor's app is limited to what the vendor has built, and changes depend on the vendor's roadmap. A custom app is limited to what you specify.&lt;/p&gt;

&lt;p&gt;If your business runs Odoo, see &lt;a href="https://treevoxa.com/services/odoo-development/" rel="noopener noreferrer"&gt;Odoo development&lt;/a&gt; before you decide. The trade-offs depend on how far the Odoo system has already been customised. For the wider comparison, read &lt;a href="https://treevoxa.com/guides/custom-erp-vs-off-the-shelf-erp-small-business/" rel="noopener noreferrer"&gt;custom ERP vs off-the-shelf ERP for small business&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  7How do you handle logins, offline use and conflicting edits?
&lt;/h2&gt;

&lt;p&gt;These three problems need a decision before design starts. Here is how we approach each.&lt;/p&gt;

&lt;h3&gt;
  
  
  Can users log in with their existing website accounts?
&lt;/h3&gt;

&lt;p&gt;Both apps should use the same backend API and the same user accounts. One login means one set of permissions. Often yes, if the accounts and their permissions were designed for it. If customers and staff sit in two separate login systems, merging them is a design job to finish before the app is built.&lt;/p&gt;

&lt;p&gt;A common pattern is OAuth 2.0 with refresh tokens. The app holds a short-lived access token and a longer-lived refresh token, so staff stay signed in without a password stored on the phone.&lt;/p&gt;

&lt;p&gt;Three things need design:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Where tokens sit on the device. Use the phone's secure storage, not a plain file.&lt;/li&gt;
&lt;li&gt;How to sign out a lost phone remotely.&lt;/li&gt;
&lt;li&gt;How role-based permissions carry across both apps, so a driver sees only their own deliveries on the web and on the phone.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  How does offline sync handle conflicting edits?
&lt;/h3&gt;

&lt;p&gt;Yes, a mobile app can work offline and sync changes later, but only for the tasks you choose. Start by deciding which tasks must work with no signal. A stock count in a steel-racked warehouse may need it. An approval may not. Design offline sync only for the tasks that need it, because sync is the hardest part to get right.&lt;/p&gt;

&lt;p&gt;Then set conflict rules by record type:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Low-risk fields&lt;/strong&gt; , such as a customer's phone number, take the latest edit.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Orders and stock&lt;/strong&gt; need a version check. The server rejects an update if the record changed after the device last read it, and the user sees both versions.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Imagine a driver in Thornhill who marks a delivery complete at 2 pm while offline. At 2:10 pm, the office changes that order's delivery address. When the phone reconnects, the server sees that the order changed after the driver last read it.&lt;/p&gt;

&lt;p&gt;The completion is not silently overwritten or dropped. The app holds it and flags the conflict. A dispatcher sees both changes and confirms the right one. Because the rule was agreed before build, the driver and the dispatcher both know what will happen.&lt;/p&gt;

&lt;h2&gt;
  
  
  8How do you roll it out without stopping the business?
&lt;/h2&gt;

&lt;p&gt;Roll out in stages, not all at once.&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Start with one module.&lt;/strong&gt; Use only the task set chosen for version one.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Choose a pilot team.&lt;/strong&gt; Pick a small group who use the process daily and will say plainly what breaks.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Run old and new side by side&lt;/strong&gt; for a period agreed in advance. Two weeks is an example. The client sets the period, and seasonal work may need longer.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Keep a rollback plan.&lt;/strong&gt; Keep the old form or process available, with a named person who can switch back within a day.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;A staged rollout suits small and mid-sized firms. A failed pilot affects one team, not the whole operation, and you get real usage data before you fund the next phase.&lt;/p&gt;

&lt;h2&gt;
  
  
  9What should you check before you hire a partner?
&lt;/h2&gt;

&lt;p&gt;Whoever builds the app will work with your ERP's API and your staff every day. Check these points before you sign.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;What to check&lt;/th&gt;
&lt;th&gt;What good looks like&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Written scope&lt;/td&gt;
&lt;td&gt;A document listing the modules, screens, endpoints and user roles included, and what is excluded&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Fixed price&lt;/td&gt;
&lt;td&gt;One agreed figure for the scoped work, with the scope and price fixed in writing before work starts&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Launch date&lt;/td&gt;
&lt;td&gt;A date agreed before work starts, tied to the agreed scope&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Code and account ownership&lt;/td&gt;
&lt;td&gt;You own the source code, app store accounts and cloud accounts from handover&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;One team for software, design and cloud&lt;/td&gt;
&lt;td&gt;One partner handles the API, the app, the design and the hosting, so nothing falls between suppliers&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;API documentation&lt;/td&gt;
&lt;td&gt;Written endpoints, authentication rules and error codes, delivered with the project&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Named technical lead&lt;/td&gt;
&lt;td&gt;One engineer is accountable for the architecture and answers your questions&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;For more on vetting suppliers, read &lt;a href="https://treevoxa.com/guides/how-to-choose-a-custom-software-development-company/" rel="noopener noreferrer"&gt;how to choose a custom software development company&lt;/a&gt;. If you want to talk through your system, &lt;a href="https://treevoxa.com/contact/" rel="noopener noreferrer"&gt;contact Treevoxa&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  10Where should you start?
&lt;/h2&gt;

&lt;p&gt;If you are working out how to add a mobile app to an existing ERP or web system, start with the endpoint audit. The task list, the endpoints behind it and the record-ownership decisions are inputs any partner will need, and you can write them yourself. Once they exist, you will know which integration route fits your system and whether adding a mobile app is a small change or a larger build.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Originally published at &lt;a href="https://treevoxa.com/guides/add-mobile-app-existing-erp-web-system/" rel="noopener noreferrer"&gt;treevoxa.com&lt;/a&gt;.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>mobile</category>
      <category>reactnative</category>
      <category>api</category>
      <category>erp</category>
    </item>
  </channel>
</rss>
