Skip to content
Koderead

Transformation Gallery

Seven systems you can open.

Each one starts from a real operational problem, follows a mapped workflow, and ends in a console your team would actually work in.

Built by Koderead to show the operating system. Figures are a worked example.

Residential estate agency

AI Sales Desk

“We pay commission-earning agents to do administrative work.”

A mid-sized agency taking enquiries from its own site, three portals, WhatsApp and the phone. Volume peaks on weekday evenings and Sunday afternoons — exactly when the office is closed. Agents spend their mornings establishing budget and timeline on enquiries that are already two days cold.

Open AI Sales Desk

Inbox

J. HaleBranch lead

Lead queue — AI Sales Desk
EnquiryRequirementSignalAI scoreRouting
S. Whitfield4-bed · Didsbury · £610kChain-free · mortgage in principle
92
Qualified
R. Okonjo3-bed · Chorlton · £425kFirst-time buyer · 1–3 months
84
Qualified
D. Laurent2-bed apt · City centre · £300kInvestor · cash purchase
79
Qualified
M. Alvarez5-bed · Altrincham · £900kNeeds to sell first · chain unclear
58
Agent review
J. PruittUnspecified · £180kBudget below area minimum
24
Nurture
Enquiries per month
1,200
Agents
14
Channels
6
First response, before
7h 40m

B2B distribution

AI Quote Engine

“We lose orders because we quote on Thursday for an enquiry that arrived Monday.”

A distributor receiving 500+ specification enquiries a week by email and PDF. Producing each quote means checking stock, applying the account's negotiated pricing tier, calculating freight to destination and confirming lead time — four systems, one salesperson, and a margin nobody verifies under deadline pressure.

Open AI Quote Engine

QEQuote Engine

M. VossSales

Parsed from enquiry

Product
XYZ-240 flange assembly
Quantity
250 units
Destination
Rotterdam, NL
Required by
15 October
Account
Existing · Tier 3

Ref QE-48291 · Rotterdam Marine Fitters BV

Assembled quote

Unit price
€184.00
Tier 3 discount
−7.0%
Freight to Rotterdam
€420.00
Line total
€43,200.00
Resulting margin
28.4%

System recommendation

Within this account's historical purchasing band and above the 24% margin floor. Stock covers 250 units from Rotterdam depot with 4 days' lead time.

  • Margin above target
  • Stock confirmed
  • Lead time meets required date
Approve & sendEditReject

Human approval is required before any quote leaves the system.

Enquiries per week
500+
Active SKUs
18,400
Pricing tiers
9
Quote turnaround, before
2.4 days

Accountancy practice

AI Finance Operations

“My qualified team spends its week processing documents instead of advising clients.”

A practice handling roughly 1,800 client documents a month across photographs, scans, portal exports and email attachments. Qualified staff transcribe rather than advise, month-end compresses against filing deadlines, and the advisory conversation that clients would actually pay for keeps getting postponed.

Open AI Finance Operations

Inbox

A. ChenPractice

Document queue — AI Finance Operations
ReferenceSupplierAmountStateValidation
INV-88214Northgate Utilities£1,284.00PostedAll checks passed
INV-88215Calder Office Supply£312.40PostedAll checks passed
INV-88216Whitrow Logistics£4,905.12ExceptionPO total differs by £68.00
INV-88217Bramley Print£196.00PostedAll checks passed
INV-88218Unrecognised sender£2,400.00ReviewSupplier not on file
INV-88219Harlow Facilities£880.50ExceptionVAT inconsistent with lines
Documents per month
1,842
Client entities
310
Qualified staff
11
Straight-through, before
0%

Regional carrier

AI Consignment Desk

“My three best operations people spend the day reading tracking numbers aloud.”

A regional carrier running roughly 400 deliveries a day. Three operations staff are nominally responsible for exception handling — failed deliveries, damage claims, escalations — but the phone rings continuously with status enquiries, so exception work happens in the gaps and therefore late. Delays are usually discovered by the customer before the carrier notices them.

Open AI Consignment Desk

Live board

Night opsShift lead

Consignment board — AI Consignment Desk
ConsignmentDestinationStatusWindowStateHandled by
CD-884120Leeds LS2Out for delivery14:20–16:00On trackSystem
CD-884121Sheffield S1At depotTomorrow 09:00–11:00On trackSystem
CD-884122Hull HU3Delayed — vehicle faultRevised 17:30–19:00DelayedSystem
CD-884123York YO1DeliveredSigned 11:42On trackSystem
CD-884124Bradford BD1Refused at doorAwaiting instructionExceptionOperations
CD-884125Wakefield WF1Address not foundAwaiting instructionExceptionOperations
Deliveries per day
400
Operations staff
3
Status calls per 100
22
Delay detection, before
Customer first

Commercial law firm

AI Precedent Search

“We keep paying twice for analysis we have already done.”

A commercial practice whose accumulated expertise sits in several thousand matter files that are, in any practical sense, unsearchable. A fee earner asks a colleague whether the firm has handled something similar; the colleague half-remembers a matter from years ago; finding it means trawling the document system by client name. Frequently the work is simply redone.

Open AI Precedent Search

PRPrecedent

Search the authorised corpus

AssociatesRead only

Query

Have we advised on a change-of-control clause triggered by an internal group reorganisation?

  1. M-2291 · Harlow Group

    Corporate · March 202494
    …the clause was held not to be triggered where the ultimate beneficial ownership remained unchanged following the intra-group transfer…

  2. M-1877 · Calder Holdings

    Corporate · November 202288
    …advised that the drafting was wide enough to capture a reorganisation, and recommended an express carve-out for transfers within the same group…

  3. M-3014 · Northgate Partners

    Commercial · July 202571
    …counterparty consent was sought on a without-prejudice basis notwithstanding the carve-out, to avoid argument at a later financing round…

Every result returns the source passage, matter and date. The system retrieves and cites; it never summarises into a conclusion and never advises.

Indexed matters
6,240
Fee earners
38
Practice areas
7
Reuse rate, before
Unmeasured

Multi-site healthcare

AI Patient Operations

“We cannot add clinical hours because reception is the bottleneck.”

A four-site clinic group where the binding constraint on capacity is administrative, not clinical. Reception handles booking, rescheduling, confirmations and a steady stream of non-clinical questions while the phone keeps ringing. Did-not-attend rates leave slots empty that a waiting list could have filled, and patients supply the same information at every touchpoint.

Open AI Patient Operations

Today

ReceptionNorthgate

Appointment board — AI Patient Operations
TimeSiteClinicianStateHow it was filled
09:00NorthgateDr A. FenwickBookedRebooked by patient, 22:40 last night
09:20NorthgateDr A. FenwickBackfilledCancellation offered to waiting list, accepted in 4m
09:40RiversideDr M. OseiBookedIntake completed pre-arrival
10:00RiversideDr M. OseiEscalatedMessage mentioned chest pain — routed to staff immediately
10:20EastwoodMs R. KaurOpenOffered to waiting list, no response yet
10:40EastwoodMs R. KaurBackfilledFilled from list by clinical priority

The 10:00 row is the important one. A message mentioning a symptom is never resolved by the system — it routes to qualified staff immediately, and the slot is held.

Appointments per week
1,960
Sites
4
Reception staff
7
Did-not-attend, before
11.4%

Direct-to-consumer retail

AI Growth Console

“Our merchandising decisions are weekly. Our data is weekly. Our competitors are daily.”

A direct-to-consumer retailer where margin, returns and stock risk are all knowable and none of them are visible until a Monday export. Abandoned baskets are chased with one template regardless of why they were abandoned, and the marketing team spends its week assembling reports rather than acting on them.

Open AI Growth Console

GRGrowth

MerchRetail

Abandonment, classified by reason

Basket-abandonment breakdown — AI Growth Console
ReasonBasketsRecoveredAction taken
Shipping cost at checkout2,140486Threshold reminder + free-shipping tier shown
Payment failure890401Retry link, alternative method offered
Size or fit hesitation1,630208Fit guidance + free returns surfaced
Comparison shopping1,18072No discount sent — margin protected
Out of stock variant57217Back-in-stock notification offered

1,184 recovered in total. Note the comparison-shopping row: the system deliberately sends no discount there, because that cohort converts anyway and a discount only gives away margin.

Flagged for a merchandiser

  • LM-4471 Linen Shirt

    Return rate 31% vs 9% category

  • LM-2210 Canvas Tote

    Margin 11.2%, below 18% floor

  • LM-8802 Wool Scarf

    14 days cover at current velocity

Asked in plain language

“Which SKUs lost us money last month?”

→ 3 SKUs below margin floor · query and source shown with the answer

Orders per month
24,800
Active SKUs
3,140
Marketing team
5
Reporting lag, before
6 days