Illustrated Kolkata development studio with a software code editor, business dashboard and mobile app prototype

CESAWS · Web, software & mobile apps

Website & Software Development Company in Kolkata

We design business websites and online stores, build custom software and Android and iOS apps, and help customers find you through SEO. Work with our Kolkata team from anywhere in India or abroad.

How your project moves from brief to launch

From the first conversation to handover, we agree on what to build, who reviews it and how to test it.

Laptop and materials on a studio workbenchShare your project brief
  1. Start with the business problem

    Tell us what users need to do, where the current process breaks down and what a successful outcome looks like. Share your current website, workflow or project notes if you have them.

  2. Agree on the scope

    We work through the priority pages or workflows, who owns the content and data, and which integrations matter. The proposal can then identify deliverables, responsibilities and review milestones.

  3. Review a working version

    Review key designs or software flows on staging. Test the journey your customers or staff will actually use, including mobile layouts, form delivery and permissions where relevant.

  4. Prepare for launch and handover

    Agree on hosting access, analytics, handover documentation and support. For redesigns, inventory important URLs and plan redirects before switching the live site.

Where we work

From Kolkata to your market.

Find web development, software and SEO services shaped around your city, audience and project requirements.

View all cities and countries

Project examples

What we can help you build

Website with a catalogue

Organise products into useful categories. Give customers a clear route from specifications to a quote or checkout. Make routine content updates available to your team.

Internal approvals tool

Start with one approval path. Map access, rules, exceptions and the reports you need before connecting data sources or rolling it out across locations.

Mobile service app

Make a frequent task easier on the phones people already use. Plan connectivity, data capture, errors and the handover to the staff who receive the results.

Campaign and SEO programme

Connect landing pages, search content and conversion measurement. Review the quality of completed enquiries alongside campaign or search reports.

Questions before you start a project

Get clear on scope, cost and what to send in your first enquiry.

How much does a business website cost?

Pricing depends on the number of pages, design complexity, content migration and integrations such as payments or booking. In your enquiry, share your current website, priority pages and target launch date so we can scope the work.

Can CESAWS redesign my existing website and keep my Google visibility?

We can review the current page URLs, ranking pages and enquiry paths before proposing changes. Where URLs change, the launch plan can include redirects, internal link updates and a post-launch indexing check.

Can you build an online store that integrates with our operations?

We can scope product management, checkout, stock updates, shipping and reporting around your current workflow. Tell us what you use for payments, inventory and delivery so we can assess the integration requirements.

Do you develop custom software and mobile apps?

Yes. We scope web applications, dashboards and Android or iOS apps around user roles, approvals, notifications and reporting. Bring an example of your current process or the first task you want your users to complete.

Can you handle SEO or digital marketing for an existing website?

Yes. We can review technical SEO, service-page content, local search visibility and lead tracking. Tell us which services and locations generate the most valuable enquiries for your business.

Do you work with clients outside Kolkata and India?

Yes. Our team is based in Kolkata and can organise remote reviews and handover across India and internationally. Tell us what market you serve, your time zone and who signs off on key decisions.

What should I send in my first enquiry?

Share your current website or a short project outline, what you want to achieve, your audience and any deadline. You can also include examples you like, needed integrations, or the cities and countries you want to reach.

What to include in your brief

  • Your current website or an example of the workflow
  • The action you want customers or staff to complete
  • Your target market, deadline and integrations
Send an enquiry

Call +91 87770 45906

Published guides

Insights for planning your project

Project planning handbook

Use the chapters below to prepare for choosing a digital partner. Open the chapters that fit your project.

Discovery and scope

Project outcome

Write what a customer or staff member must be able to do after release, the present obstacle and what success looks like. Agree on one baseline and the timeframe in which you can measure a change.

A claim such as improve engagement is hard to accept. A visitor reaching the correct department with a complete enquiry is observable.

The project owner can use this part of the plan when reviewing choosing a digital partner. Check who supplies each example, who decides whether it reflects normal work and what sort of unexpected case might change the scope or release order.

Starting point for the brief: your first conversation about a project.

What to prepare · Outcome statement, baseline and measurement owner

Priority audience

Name the primary audience, its devices and what information it already has when the task begins. List secondary users separately to keep a first release coherent.

A procurement team may understand part numbers while a customer is searching by application or symptom. Useful entry points reflect both habits.

Review how a failure or delay will be detected. Decide who is responsible for the first response, which record captures the decision and what happens if a customer needs help before a fix is available. Give the receiving person a brief explanation of why the state occurred and which actions they can take safely.

What to prepare · Audience descriptions and representative tasks

Evidence from the current system

Collect examples of ordinary work, exceptions, delays and repairs. Ask someone who performs the task to explain what the record means and where the official version lives.

Scattered records often contain useful context despite inconsistent labels. Test a redacted extract before relying on a clean-looking sample made for a meeting.

At handover, ask the owner to repeat an ordinary update while another reviewer observes. This exposes permissions, missing context and steps that depend on a specific person or private account. Retain the example in the project notes. Check that a new team member can identify the source of the existing decision and request a change.

What to prepare · Representative cases and source inventory

Release boundaries

Separate the first useful outcome from enhancements that can wait. Identify requirements with dependencies that are late, expensive or outside your organisation’s control.

Agree whether a pilot can accept manual follow-up before automating secondary stages. State how the handoff will work during that temporary period.

For a release with several markets, branches or customer groups, check which parts of this decision apply to everyone and which require a local choice. Document the rule and a representative example where possible. Ask the appropriate regional or operational reviewer to confirm whether the result can still be used in their normal working conditions.

What to prepare · First release scope and separation of future work

Acceptance criteria

Write the starting condition, the action and the expected result for each important scenario. Include what happens to the record and what the responsible person sees.

A form test should confirm the enquiry arrives in the right queue with its original context. A confirmation message alone is incomplete evidence.

Record what will be measured at launch and later during ordinary use. Choose enough examples to make a comparison worthwhile, and agree on an action if the result falls outside the expected range. Share which variations the team can change without a new development milestone or an update to the published customer journey.

What to prepare · Scenario list with observable expected results

Dependency decisions

List external providers, approvals, access, content and data that affect delivery. Distinguish prerequisites from tasks that can proceed independently.

Record who can obtain a test account, what settings differ from production and how support should respond when a provider becomes unavailable.

If several options look useful, evaluate each using the same example. Compare what the user needs to do, what the operating team has to maintain and how easy it is to reverse an incorrect decision. Capture the reason for the preferred option while the competing alternatives and practical constraints are still clear.

What to prepare · Dependency register with deadlines and owners

Estimate and change requests

Link the estimate to working assumptions and review milestones. Track later requests by the problem they address, the effect on users and the evidence needed to approve the change.

Schedule changes that affect more than one process after checking connected interfaces and operational handovers. Keep open decisions visible to both teams.

Decide which portion of this work belongs in the first release. Name a risk that can be observed and a condition that would require waiting. Note which work can progress in parallel and which input needs approval. At the next review, compare the expected outcome with the actual evidence and record the changes in the scope.

What to prepare · Scope log and a documented review point

Experience and content

Information architecture

Observe how people group information before designing navigation. Use page and screen labels that match real customer questions, then test whether they support a realistic task.

Avoid relying exclusively on internal department names. A customer reading on mobile should understand where to go without opening every menu.

When mapping choosing a digital partner, choose a boundary that the team recognises. Describe what enters the work, what leaves it and who receives the result. Note any existing agreement, published information or timed dependency that still requires confirmation.

What to prepare · Page tree and task-based navigation checks

Mobile starting point

Sketch the smallest supported view first. Review when primary actions are visible, whether content wraps correctly and whether controls are easy to reach.

Check long names, magnified text and keyboards on actual devices. Desktop designs that scale down well can still hide an important prompt below the fold.

Add this item to the project plan as an observable step. Record the person or team who supplies the input, the way you will verify it and any other milestone that must finish first. If work is split across teams, decide what information can be shared early and document the decision that is still outstanding.

What to prepare · Responsive layouts and small-screen acceptance checks

Form design

Request the minimum information needed to direct or complete the task. Set clear labels, indicate required fields and explain how an unsuccessful submission can be corrected.

Use input types and keyboard suggestions that suit the field. Preserve entered information where appropriate when a connection or validation check fails.

A useful test includes the condition that starts the work, a single action and an outcome that a reviewer can actually observe. Note whether the result is visible to a customer, an administrator or both. Keep an example that support can repeat after a release so an unexpected change can be compared with the original acceptance result.

What to prepare · Field list, validation rules and confirmation copy

Accessible UI

Review keyboard order, visible focus, plain labels, contrast and text resizing. Check important tasks with assistive technology where your project supports it.

Account for disabled, loading, expired, failed and completed states. Users need a way to recognise each result without depending on colour or position alone.

Invite the people who will use or maintain this part of the service to the review. Give them a realistic brief, ask them to complete the intended task and note the point where they ask a question. Decide whether the answer belongs in the interface, the operating procedure or a separate support note, and name who owns that change.

What to prepare · Accessibility review and testable component states

Design systems

Define typography, spacing, colour, content hierarchy and interaction patterns alongside variants of common parts. Explain when exceptions are appropriate.

A component has to support longer text, unfamiliar data and a narrow device. Write design notes around the choices a developer or editor will repeat.

Write down the assumption that would be most expensive if it turned out to be wrong. Consider how you could verify it in a small sample before the full scope is approved. Set a deadline for resolving the assumption and explain what happens to work that is already under way if the dependency cannot be confirmed.

What to prepare · Style tokens, component examples and implementation notes

Editorial ownership

Assign responsibility for product copy, navigation labels, translation and imagery. Record what an editor can update safely after launch.

Use a checklist when publishing important pages. Require a clear main heading, useful alternative text, a relevant next step and a named reviewer.

Review how a failure or delay will be detected. Decide who is responsible for the first response, which record captures the decision and what happens if a customer needs help before a fix is available. Give the receiving person a brief explanation of why the state occurred and which actions they can take safely.

What to prepare · Content matrix, review workflow and editor training

Brand and image preparation

Gather approved logos, original assets, usage rights, output sizes and a concise set of visual rules. Choose crops based on where images appear and what they explain.

Use realistic replacement material if the client cannot share an internal screen. Mark recommendations and concept examples clearly when showing them as work.

At handover, ask the owner to repeat an ordinary update while another reviewer observes. This exposes permissions, missing context and steps that depend on a specific person or private account. Retain the example in the project notes. Check that a new team member can identify the source of the existing decision and request a change.

What to prepare · Asset inventory with permissions and export specifications

International experience

Write down which language, contact, currency and delivery details change by market. Decide which values can be configured and which require dedicated copy.

A translation should survive real text lengths and reading direction. Test an actual translated example before setting fixed sizes or approving a single mockup.

For a release with several markets, branches or customer groups, check which parts of this decision apply to everyone and which require a local choice. Document the rule and a representative example where possible. Ask the appropriate regional or operational reviewer to confirm whether the result can still be used in their normal working conditions.

What to prepare · Language and market matrix with review owners

Software and operations

Roles and access

Define who creates, views, edits, approves and exports each kind of information. Review rules for multi-branch work, temporary staff and transferred ownership.

Show both authorised and unauthorised states during acceptance. A shortcut that reveals restricted information can undermine an otherwise usable workflow.

In the context of choosing a digital partner, keep one representative example available throughout delivery. Use that example to answer questions about the design, demonstrate the intended changes and confirm that the final handover covers the way staff will handle common exceptions.

What to prepare · Role matrix and approved access scenarios

Workflow states

Give each state a definition and name the actions that move a record to the next state. Decide who can cancel, return or reopen a request.

Keep reasons for manual exceptions and distinguish a temporary delay from a final decision. Ask report owners how those events should appear in totals.

Decide which portion of this work belongs in the first release. Name a risk that can be observed and a condition that would require waiting. Note which work can progress in parallel and which input needs approval. At the next review, compare the expected outcome with the actual evidence and record the changes in the scope.

What to prepare · Workflow map and transition rules

System of record

Identify which system owns the current value of a product, customer, order or task. Set boundaries for fields that can be copied and where staff correct a mistake.

If two systems can change a value, define what happens when they disagree and how an operator recognises a mismatch.

Before approving this part of the project, choose a person who can explain the operational result. Ask that person to demonstrate the typical path and a case where the input is late, incomplete or outside the expected range. Record what the business considers an acceptable result and agree who can decide when there is a disagreement.

What to prepare · Data ownership matrix and correction rules

Migration and cleanup

Measure the source quality and define how to map legacy codes, duplicates, old attachments and incomplete records. Test with a copy before scheduling the switch.

Use counts and a small agreed sample to verify results. Retain the source export and decide what can be corrected after the first release.

Use the first review to collect a concrete example rather than only an approval. Photograph or describe the previous result where appropriate, then compare it with the proposed one. Ask the people who depend on the output to explain what they will do differently once the change is available. Update the open questions while the example is still easy to reproduce.

What to prepare · Mapping file, migration report and recovery checkpoint

API handoffs

Write a request and response contract for each exchange. Include validation, retry behaviour, ownership of credentials and an observable result when a request fails.

Use examples of missing fields and a slow or delayed provider. Confirm a repeated request can be recognised before creating a duplicate downstream record.

Add this item to the project plan as an observable step. Record the person or team who supplies the input, the way you will verify it and any other milestone that must finish first. If work is split across teams, decide what information can be shared early and document the decision that is still outstanding.

What to prepare · Integration contract and test payloads

Reporting and exports

Define what each report counts, which stages are included and how to use time ranges. Check exports against source records with an operations reviewer.

A team might label a quote as completed even though the customer has yet to accept it. Use labels that agree with the actual workflow state.

A useful test includes the condition that starts the work, a single action and an outcome that a reviewer can actually observe. Note whether the result is visible to a customer, an administrator or both. Keep an example that support can repeat after a release so an unexpected change can be compared with the original acceptance result.

What to prepare · Reporting definitions with sample reconciliations

Human review and automation

Identify automatic decisions, items that need approval and the conditions that send a case to a manual queue. Record what an operator can recover without developer help.

Limit the first automation to a small set of reliable inputs. Compare ordinary and unusual cases and document a useful fallback.

Invite the people who will use or maintain this part of the service to the review. Give them a realistic brief, ask them to complete the intended task and note the point where they ask a question. Decide whether the answer belongs in the interface, the operating procedure or a separate support note, and name who owns that change.

What to prepare · Review queue, escalation path and quality examples

Support and operational logs

Log enough context to resolve a case and correlate it with a record or request. Clarify which operational data can be visible to which team.

A support reviewer should be able to tell whether an integration accepted an update, retrying makes sense and who can make a correction.

Write down the assumption that would be most expensive if it turned out to be wrong. Consider how you could verify it in a small sample before the full scope is approved. Set a deadline for resolving the assumption and explain what happens to work that is already under way if the dependency cannot be confirmed.

What to prepare · Support notes, error ownership and audit trail

Ecommerce and lead generation

Catalogue structure

Agree a stable item identifier, categories, variants, availability and which attributes buyers need to compare. Ask the catalogue owner to demonstrate a realistic update.

A variant can inherit information from its parent, but buyers still need to see the details that change price, delivery or suitability.

For choosing a digital partner, use the project brief to identify where this decision belongs. Tie it to an example the business recognises and make sure the people involved in the review agree what the evidence will show.

What to prepare · Catalogue model and sample entries

Quoting and checkout

Map the actions a buyer should take after choosing a product or service. Identify minimum order quantities, discounts, delivery coverage and payment states when applicable.

Test the difference between a request for a quote and a confirmed purchase. Set expectations about what the customer receives and who responds next.

For a release with several markets, branches or customer groups, check which parts of this decision apply to everyone and which require a local choice. Document the rule and a representative example where possible. Ask the appropriate regional or operational reviewer to confirm whether the result can still be used in their normal working conditions.

What to prepare · Buyer journey and acceptance tests for each outcome

Enquiry qualification

Agree which information makes a request useful to the team and who owns a lead after it arrives. Ask for fields that help with assignment or follow-up.

Keep a clear route for a visitor who is unsure which service applies. Avoid inventing a forced classification that delays their response.

Record what will be measured at launch and later during ordinary use. Choose enough examples to make a comparison worthwhile, and agree on an action if the result falls outside the expected range. Share which variations the team can change without a new development milestone or an update to the published customer journey.

What to prepare · Qualification criteria and routing rules

Returns and exceptions

Write down what happens when stock changes, a transaction is cancelled or a service cannot be delivered as requested. Decide which status the customer and staff can each see.

Test a partial return and a delayed acknowledgement with sample records. The support team should be able to confirm the outcome without reconstructing it from messages.

If several options look useful, evaluate each using the same example. Compare what the user needs to do, what the operating team has to maintain and how easy it is to reverse an incorrect decision. Capture the reason for the preferred option while the competing alternatives and practical constraints are still clear.

What to prepare · Exception table and support handover

Conversion measurement

Define a completed action and verify that it is counted once. Separate the first user interaction, a submitted request and an outcome reviewed by staff.

Test the reporting with known examples and identify which data is unavailable when a visitor denies optional tracking. Keep reporting definitions visible.

Decide which portion of this work belongs in the first release. Name a risk that can be observed and a condition that would require waiting. Note which work can progress in parallel and which input needs approval. At the next review, compare the expected outcome with the actual evidence and record the changes in the scope.

What to prepare · Event map and verified conversion examples

Landing pages and campaigns

Connect the ad message or referral source to the page that answers the visitor’s immediate question. Explain what the business offers and where it can deliver.

Check mobile load time, the offer, the enquiry route and how staff classify a request. Compare the quality of leads before expanding campaign spend.

Before approving this part of the project, choose a person who can explain the operational result. Ask that person to demonstrate the typical path and a case where the input is late, incomplete or outside the expected range. Record what the business considers an acceptable result and agree who can decide when there is a disagreement.

What to prepare · Campaign brief, landing page checklist and review cadence

Follow-up workflow

Decide who can assign, respond to and close a request. Capture a short status history so the customer can receive consistent replies from multiple people.

Define when to merge contacts and when to retain separate requests. An enquiry for a different product may belong to the same customer but need a different reviewer.

Use the first review to collect a concrete example rather than only an approval. Photograph or describe the previous result where appropriate, then compare it with the proposed one. Ask the people who depend on the output to explain what they will do differently once the change is available. Update the open questions while the example is still easy to reproduce.

What to prepare · Response workflow and status definitions

Search and publication

Search intent

Group queries around the question a visitor needs answered. Put buyer-facing comparisons, commercial service information and how-to material on pages suited to their purpose.

Review results for the terms people already use and compare the landing pages with your offering. Distinct intent is a stronger publishing reason than a slight phrasing change.

When evaluating choosing a digital partner, ask which details affect the way the business receives or acts on information. Choose a testable example and note whether the project needs new content, new configuration or a change in the process.

What to prepare · Query groups, preferred pages and update priorities

Title and description

Write a specific title and a useful description representing the actual page. Put the service, question or product near the beginning where it fits naturally.

Check that different page titles describe their own purpose. A visitor should recognise the answer to their query before opening the listing.

Invite the people who will use or maintain this part of the service to the review. Give them a realistic brief, ask them to complete the intended task and note the point where they ask a question. Decide whether the answer belongs in the interface, the operating procedure or a separate support note, and name who owns that change.

What to prepare · Page metadata and a duplicate review

Canonical and redirects

Choose the preferred address for a page and use it consistently in internal links and the sitemap. Map retired URLs to the nearest useful replacement.

Test common old addresses before launch. When a source served a distinct purpose, avoid sending all visitors to a generic page without explanation.

Write down the assumption that would be most expensive if it turned out to be wrong. Consider how you could verify it in a small sample before the full scope is approved. Set a deadline for resolving the assumption and explain what happens to work that is already under way if the dependency cannot be confirmed.

What to prepare · Canonical policy and redirect map

Crawl paths

Make important pages reachable through ordinary links. Give a link meaningful text and use a page hierarchy that visitors can browse without guessing.

A directory, service page, related guide and market page can support one another when each serves a useful purpose. Avoid adding repeated link blocks everywhere.

Review how a failure or delay will be detected. Decide who is responsible for the first response, which record captures the decision and what happens if a customer needs help before a fix is available. Give the receiving person a brief explanation of why the state occurred and which actions they can take safely.

What to prepare · Navigation, breadcrumbs and internal-link review

Structured data

Describe the page, organisation, service, article or breadcrumbs using information shown in the site. Validate fields and URLs after the final design is published.

Use business locations and claims that can be verified. Keep paid offerings and examples clearly described so marked-up data matches the visible page.

At handover, ask the owner to repeat an ordinary update while another reviewer observes. This exposes permissions, missing context and steps that depend on a specific person or private account. Retain the example in the project notes. Check that a new team member can identify the source of the existing decision and request a change.

What to prepare · Structured-data output and validation results

Sitemap and indexing

Generate a sitemap from preferred public URLs that return usable content. Review indexability and submit the production version through the owner’s Search Console account.

For old pages, consider the existing visits, links and purpose before removing them. Look at page groups when investigating unexpected exclusion.

For a release with several markets, branches or customer groups, check which parts of this decision apply to everyone and which require a local choice. Document the rule and a representative example where possible. Ask the appropriate regional or operational reviewer to confirm whether the result can still be used in their normal working conditions.

What to prepare · Sitemap audit and index coverage review

Performance of media

Export assets for the display sizes they actually occupy and specify image dimensions so the layout remains stable. Load the leading visual earlier than supporting images.

Check download size and JPEG decoding, mobile rendering and what the page looks like if an image request fails. Verify that meaningful alternative text remains useful.

Record what will be measured at launch and later during ordinary use. Choose enough examples to make a comparison worthwhile, and agree on an action if the result falls outside the expected range. Share which variations the team can change without a new development milestone or an update to the published customer journey.

What to prepare · Image manifest, viewport review and performance baseline

Results and iteration

Track impressions, relevant visits and completed customer actions using the pages and queries that matter to the business. Keep a record of publication changes.

Review results after a meaningful period. A report that distinguishes indexability, usability and demand can explain more than a list of positions alone.

If several options look useful, evaluate each using the same example. Compare what the user needs to do, what the operating team has to maintain and how easy it is to reverse an incorrect decision. Capture the reason for the preferred option while the competing alternatives and practical constraints are still clear.

What to prepare · Baseline report and review schedule

Quality and release

Browser and device checks

Test realistic viewport widths, long input, text enlargement, navigation, forms and images. Check that dominant actions remain within reach on smaller devices.

Write the expected result and capture any visual or functional defect that affects the task. Recheck fixes on the device where you found them.

Discuss how this applies to choosing a digital partner before estimating the full work. If a dependency prevents you from demonstrating the outcome today, record who can provide the missing material and what can be reviewed in the meantime.

What to prepare · Device list, behaviour checks and defect evidence

Staging environment

Review the first release with realistic but redacted data, separate credentials and representative deployment settings. Identify any deliberate differences from production.

Test the settings that affect links, secure forms, media and other expected outcomes. Agree who can move work into production after reviewing the results.

Use the first review to collect a concrete example rather than only an approval. Photograph or describe the previous result where appropriate, then compare it with the proposed one. Ask the people who depend on the output to explain what they will do differently once the change is available. Update the open questions while the example is still easy to reproduce.

What to prepare · Staging checklist and approval record

Failure scenarios

Plan checks for incomplete input, cancelled actions, a repeated submission and interrupted connectivity. Use cases that belong to the intended product.

Test what the visitor sees and what the receiving team must act on. An error message should leave enough information to allow the person to recover.

Add this item to the project plan as an observable step. Record the person or team who supplies the input, the way you will verify it and any other milestone that must finish first. If work is split across teams, decide what information can be shared early and document the decision that is still outstanding.

What to prepare · Failure tests with expected customer and staff behaviour

Credential ownership

Identify who owns and can recover project accounts. Keep operational access in approved channels and grant only the permissions needed for the agreed tasks.

Include ownership for the hosting account, domain, deployment, analytics and external integrations. Review access after a contractor or staff departure.

A useful test includes the condition that starts the work, a single action and an outcome that a reviewer can actually observe. Note whether the result is visible to a customer, an administrator or both. Keep an example that support can repeat after a release so an unexpected change can be compared with the original acceptance result.

What to prepare · Account inventory and handover responsibilities

Release and rollback

Identify the point where switching traffic, configuring a provider or moving data becomes visible to users. Define when to hold or reverse a release.

Rehearse how you can restore a previous configuration and verify the essential task afterward. Decide how to communicate delays to the people affected.

Invite the people who will use or maintain this part of the service to the review. Give them a realistic brief, ask them to complete the intended task and note the point where they ask a question. Decide whether the answer belongs in the interface, the operating procedure or a separate support note, and name who owns that change.

What to prepare · Release plan with rollback criteria

Backups and restore drills

Check what is backed up, how often it is recoverable and which environments or assets require a separate copy. Test a restore to a safe location.

Do not assume source code will include a database, configuration or uploaded media. Verify the pieces needed to resume service after reverting a release.

Write down the assumption that would be most expensive if it turned out to be wrong. Consider how you could verify it in a small sample before the full scope is approved. Set a deadline for resolving the assumption and explain what happens to work that is already under way if the dependency cannot be confirmed.

What to prepare · Backup manifest and restore evidence

Post-release review

Choose a small list of essential tasks and measurements to review once work is public. Name who will receive issues and who can approve a correction.

Compare results to the staging baseline. Identify current bugs, usability improvements and new requests so immediate support stays focused.

Review how a failure or delay will be detected. Decide who is responsible for the first response, which record captures the decision and what happens if a customer needs help before a fix is available. Give the receiving person a brief explanation of why the state occurred and which actions they can take safely.

What to prepare · Launch checks and issue triage notes

Working together

Working hours and reviews

Agree the hours for decision-making and the channel for routine and urgent questions. Identify what can be approved in writing when teams work in different time zones.

Ask reviewers to capture the observable problem when a handover does not match expectations. Record decisions alongside the relevant brief or milestone.

Start the discussion of choosing a digital partner by confirming what the current method already handles well. Add the proposed change only after the team can explain the problem, identify a useful sample and compare it with the intended result.

What to prepare · Review calendar and feedback ownership

Source and documentation

Define access to source, designs, content, environment settings and support material. Show how someone new can identify where an operational change belongs.

Test documentation by asking another person to repeat a common task. If they need privileged access or help from the original developer, record the dependency.

Record what will be measured at launch and later during ordinary use. Choose enough examples to make a comparison worthwhile, and agree on an action if the result falls outside the expected range. Share which variations the team can change without a new development milestone or an update to the published customer journey.

What to prepare · Ownership and handover checklist

Training and editorial handover

Decide which regular updates staff will perform and which tasks require a deployment or more specialised support. Use examples drawn from your real work.

During training, ask someone to correct a mistake rather than only performing the happy path. Give them a named route for asking about unusual cases.

If several options look useful, evaluate each using the same example. Compare what the user needs to do, what the operating team has to maintain and how easy it is to reverse an incorrect decision. Capture the reason for the preferred option while the competing alternatives and practical constraints are still clear.

What to prepare · Training exercises and support scope

Build versus integrate

Compare the recurring cost and operating burden of a build with the fit of an existing product. Check whether the provider exposes the interfaces and data you need.

Use one complete sample journey to compare alternatives. Consider how records can be exported and what happens if a provider changes its offering.

Decide which portion of this work belongs in the first release. Name a risk that can be observed and a condition that would require waiting. Note which work can progress in parallel and which input needs approval. At the next review, compare the expected outcome with the actual evidence and record the changes in the scope.

What to prepare · Decision record with assumptions and trade-offs

Cost of ownership

List ongoing needs such as hosting, editorial updates, licences, integration maintenance and operational review. Identify the team that pays for and manages each part.

A quote can be evaluated more confidently when dependencies and expected reviews are clear. Ask how future estimates will be scoped and approved.

Before approving this part of the project, choose a person who can explain the operational result. Ask that person to demonstrate the typical path and a case where the input is late, incomplete or outside the expected range. Record what the business considers an acceptable result and agree who can decide when there is a disagreement.

What to prepare · Ownership cost checklist and support boundaries

Next release opportunities

Set a schedule for reviewing user feedback, performance and common support issues. Choose improvements tied to an observed need and an owner who can verify them.

Create a short backlog with what the change would improve, who benefits and how to know whether it worked. Avoid adding features that no one is ready to maintain.

Use the first review to collect a concrete example rather than only an approval. Photograph or describe the previous result where appropriate, then compare it with the proposed one. Ask the people who depend on the output to explain what they will do differently once the change is available. Update the open questions while the example is still easy to reproduce.

What to prepare · Early improvement list with success criteria

Explore the work involved in your project

Bring leads to a business website

Connect the first search or campaign visit to pages that explain the service and a clear enquiry route. Review the website development service for designs, CMS options, integrations and website redesign.

Build software for customers or staff

Identify the job people need to complete, from approving an order to making a booking. Explore custom software development and mobile apps for workflow planning, access, integrations and release testing.

Improve a digital marketing programme

Link search visibility and campaign content to qualified demand. Start with the pages and actions you can measure, then review where leads are accepted and followed up.

What would you like to build?

Tell us about the project and where you need help. Choose web design, software, mobile apps or digital marketing in the enquiry form.

Start your enquiry