<?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: Capital Complete Solutions</title>
    <description>The latest articles on DEV Community by Capital Complete Solutions (@capital-cs).</description>
    <link>https://dev.to/capital-cs</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%2F4235000%2Ff4cccd1f-9da7-4e9d-a3d8-2a567b371293.webp</url>
      <title>DEV Community: Capital Complete Solutions</title>
      <link>https://dev.to/capital-cs</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/capital-cs"/>
    <language>en</language>
    <item>
      <title>What property inventory software actually needs to do</title>
      <dc:creator>Capital Complete Solutions</dc:creator>
      <pubDate>Wed, 11 Nov 2026 07:30:53 +0000</pubDate>
      <link>https://dev.to/capital-cs/what-property-inventory-software-actually-needs-to-do-30mn</link>
      <guid>https://dev.to/capital-cs/what-property-inventory-software-actually-needs-to-do-30mn</guid>
      <description>&lt;p&gt;I am not a developer. I write property inventory reports for a living — the documents that record what condition a rented property is in before a tenant moves in, and what condition it is in when they leave. When the two do not match, somebody loses a deposit.&lt;/p&gt;

&lt;p&gt;Every few months another startup launches a tool for this, and most of them get the same things wrong. Not because the engineering is bad, but because the domain spec is sitting in the heads of people like me and nobody writes it down.&lt;/p&gt;

&lt;p&gt;So here it is. If you ever build in this space, or anything else where a record has to survive someone disagreeing with it, this is the part that matters.&lt;/p&gt;

&lt;h2&gt;
  
  
  What the job actually is
&lt;/h2&gt;

&lt;p&gt;Three visits, over months or years:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;
&lt;strong&gt;Check-in&lt;/strong&gt; — record the property's condition at the start of the tenancy.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Mid-term&lt;/strong&gt; — optional, somewhere in the middle.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Check-out&lt;/strong&gt; — record the condition at the end.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Then somebody compares visit 1 and visit 3 and decides who pays for the difference. In the UK that somebody is a deposit scheme adjudicator, the process is documentary-only, and the burden of proof sits entirely with the landlord.&lt;/p&gt;

&lt;p&gt;That last sentence is the whole specification. &lt;strong&gt;The output of the system is evidence, and evidence is read by someone who is not on your side.&lt;/strong&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Requirement 1: the diff is the product
&lt;/h2&gt;

&lt;p&gt;Most tools treat each visit as an independent document. Capture rooms, capture items, capture photos, export a PDF. Ship.&lt;/p&gt;

&lt;p&gt;That builds the easy half. A record of the end state proves nothing on its own — a photo of a damaged worktop establishes that a worktop is damaged, not when it happened. The valuable artefact is never a single report. It is the &lt;strong&gt;pair&lt;/strong&gt;, and the comparison between them.&lt;/p&gt;

&lt;p&gt;So the data model has to make "same item, two points in time" a first-class thing, not something a human reconstructs at the end:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Property
  └── Area (kitchen, rear bedroom, hallway)
        └── Item (worktop, carpet, window frame)   ← stable ID across visits
              └── Observation
                    ├── visit_id
                    ├── condition (enum, not free text)
                    ├── notes
                    └── Photo[]  ← each tied to the item, not the visit
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The stable &lt;code&gt;Item&lt;/code&gt; ID is the bit people skip. Without it you cannot diff anything, and you are back to a human reading two PDFs side by side.&lt;/p&gt;

&lt;h2&gt;
  
  
  Requirement 2: enforce structure at capture, not at export
&lt;/h2&gt;

&lt;p&gt;If the check-out visit can invent its own room names, its own item list and its own ordering, comparability is already gone — and you cannot fix it in the export layer.&lt;/p&gt;

&lt;p&gt;A check-out should be &lt;strong&gt;generated from the check-in&lt;/strong&gt;: same areas, same items, same order, pre-populated, with the clerk recording what changed. Not a blank form.&lt;/p&gt;

&lt;p&gt;This also makes the field work faster, which is the thing that actually gets a tool adopted.&lt;/p&gt;

&lt;h2&gt;
  
  
  Requirement 3: photos need identity, not storage
&lt;/h2&gt;

&lt;p&gt;Every tool stores photos. Very few give them identity.&lt;/p&gt;

&lt;p&gt;A photo that cannot be matched to the same subject in the earlier visit is decoration. What makes a pair persuasive is that it is obviously the same thing, shot the same way, before and after.&lt;/p&gt;

&lt;p&gt;Practical version:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;A photo belongs to an &lt;code&gt;Item&lt;/code&gt;, not to a visit folder.&lt;/li&gt;
&lt;li&gt;At check-out, show the clerk the check-in photo for that item and let them frame the new shot against it.&lt;/li&gt;
&lt;li&gt;Export them adjacent, not in separate appendices.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;That one UI decision — showing the previous photo while taking the new one — is worth more than any amount of image processing.&lt;/p&gt;

&lt;h2&gt;
  
  
  Requirement 4: metadata has to be automatic
&lt;/h2&gt;

&lt;p&gt;An undated photograph proves condition but not timing, and timing is almost always the thing in dispute.&lt;/p&gt;

&lt;p&gt;Two traps here:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;EXIF gets stripped.&lt;/strong&gt; Upload to most cloud storage, or let a phone share-sheet touch it, and the timestamp is gone. Capture the timestamp at the moment of capture and store it in your own record. Do not rely on the file carrying it.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Device clocks lie.&lt;/strong&gt; Stamp server-side where you can, and keep the device time too so a discrepancy is visible rather than silent.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Anything that depends on a person remembering to record something will be missing on the day it matters.&lt;/p&gt;

&lt;h2&gt;
  
  
  Requirement 5: offline-first is not optional
&lt;/h2&gt;

&lt;p&gt;Empty properties have no Wi-Fi, often no furniture, and frequently no signal — think a basement flat in a Victorian conversion with thick walls.&lt;/p&gt;

&lt;p&gt;A clerk spends 45–90 minutes in that property capturing a few hundred photos. If your app needs a connection to save, it will lose someone's morning, once, and then never be used again.&lt;/p&gt;

&lt;p&gt;Local-first storage, background sync, conflict handling that does not silently drop a visit. Standard stuff, but it is a hard requirement here, not a nice-to-have.&lt;/p&gt;

&lt;h2&gt;
  
  
  Requirement 6: the export is an argument, not an archive dump
&lt;/h2&gt;

&lt;p&gt;The deposit schemes list over-submission as a known failure mode — bundles stuffed with material that has no bearing on the actual question. When a reader has to hunt for your argument, you have made them do your work, and they will do it less generously than you would.&lt;/p&gt;

&lt;p&gt;So the export should not be "everything we captured, in capture order." It should be able to produce a &lt;strong&gt;claim bundle&lt;/strong&gt;: for each disputed item, the check-in state, the check-out state, the two photographs, and the difference, on one page.&lt;/p&gt;

&lt;p&gt;Full archive available. Argument presented.&lt;/p&gt;

&lt;h2&gt;
  
  
  What this looks like as a product decision
&lt;/h2&gt;

&lt;p&gt;If you are scoping an MVP in this space, the ranking is roughly:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Stable item identity across visits&lt;/li&gt;
&lt;li&gt;Check-out generated from check-in&lt;/li&gt;
&lt;li&gt;Photo pairing in the capture UI&lt;/li&gt;
&lt;li&gt;Automatic, trustworthy timestamps&lt;/li&gt;
&lt;li&gt;Offline capture&lt;/li&gt;
&lt;li&gt;Claim-bundle export&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Everything else — pretty PDFs, e-signatures, branding, integrations with lettings CRMs — matters commercially but does not change whether the output holds up.&lt;/p&gt;

&lt;h2&gt;
  
  
  The generalisable bit
&lt;/h2&gt;

&lt;p&gt;Strip out the property domain and this is a pattern that shows up anywhere you capture state now and compare it later: monitoring, audits, compliance records, incident reviews, before-and-after of any kind.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Capture the baseline when it is cheap and boring.&lt;/strong&gt; Keep the structure identical across captures so the comparison is mechanical. Make metadata automatic. And design the output for a reader who is sceptical, in a hurry, and only reading it once.&lt;/p&gt;

&lt;p&gt;The property version of this problem just has the unusual property that somebody loses real money, in public, when you get it wrong — which is a good way to find out which of your assumptions were decorative.&lt;/p&gt;




&lt;p&gt;&lt;em&gt;I run &lt;a href="https://www.capital-cs.com" rel="noopener noreferrer"&gt;Capital Complete Solutions&lt;/a&gt;, an AIIC-accredited property inventory and inspection company in Birmingham, UK. We produce the &lt;a href="https://www.capital-cs.com/services/inventory-reports" rel="noopener noreferrer"&gt;inventory&lt;/a&gt; and &lt;a href="https://www.capital-cs.com/services/check-out-inspections" rel="noopener noreferrer"&gt;check-out reports&lt;/a&gt; described above, currently without a tool that does all of the above — which is how I know what is missing.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;If you are building in this space and want the unglamorous domain detail, I am happy to answer questions in the comments.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>architecture</category>
      <category>startup</category>
      <category>productivity</category>
    </item>
  </channel>
</rss>
