An assistant system for The Gilchrist Collection. It handles routine questions, sheets and follow-ups so coordinators spend their time with couples, and a research lab tests new ideas for the business. Shown as three working sketches by Linards Berzins.
Checks my email, calendar, Slack and CRM every 2 hours. Over 2,000 automated checks run against it.
Bridebook, about 7,000 couples
Bridebook UK Wedding Report, April 2026
The Gilchrist Collection, CEO page
Eleven ideas, grouped by who pays. They share one asset a generic app lacks: how a venue group actually runs, which is the first thing I would learn from you. The three marked below are the ones I sketch in the next section.
Sold as package extras or included to win the booking.
Why Gilchrist: Answers come from your own guides and contracts, not the open internet.
Why Gilchrist: Only items your venues and suppliers can actually provide.
Why Gilchrist: If you hold real timings from past weddings, a generic app cannot match them.
Why Gilchrist: Guests get quick answers on the day, so the couple and the team can enjoy it.
Why Gilchrist: You are there at the moment couples care most.
Tools that help your teams, with people always in charge.
Why Gilchrist: It would use your own prices, availability and the replies that have worked, if you keep them.
Why Gilchrist: From your public sites, each venue appears to have its own brochure form and tour booking, so joining them up is a group job.
Why Gilchrist: It would draft the sheets your teams already write by hand, from the couple’s choices, for them to check, if they do.
Why Gilchrist: Built around each venue’s rooms and kitchens, with a person always signing off.
Helps Gilchrist spot what couples want next, and test it before building.
Why Gilchrist: Twelve venues and 2,500+ weddings a year are a rich source of questions, if you want them studied. Set up in October 2026, first questions running.
Only after it works across your own twelve.
Why Gilchrist: A playbook that would first be tested on your own sites.
Size is my first estimate: weeks, or a few months. It is a sketch, not a quote. Which data you hold today is the first thing I would check.
Couples get a planning companion from the day they book. Your teams get clearer requests and more time with couples. Everything is a sketch for a made-up demo venue, with pre-written answers and invented rooms and prices.
Change any number to see the effect. The 2,500+ weddings a year is from your website; every other number is an example, so swap in your own.
Not a forecast, and it ignores build and running costs.
Example: hours freed for couples a year1,667
Example: add-on revenue a year£37,500
Hours freed = weddings × questions × minutes × share answered straight away ÷ 60. Revenue = weddings × take-up × price.
Sketch: a demo venue with invented rooms and made-up prices. Not a Gilchrist venue, guide or quote.
The problem A couple wants a straight answer about their own day, without waiting for an email reply.
What I’d test first Do couples trust an answer more when it shows the page it came from, and how many questions still need a person? Not tested yet.
This sketch needs JavaScript to be tappable.
The couple’s phone
Amy and Sam. Saturday 14 November, The Glasshouse at Fernhill (demo venue).
Try asking
Sketch: it recognises a few words from the demo guide. Anything else goes to Hannah, your coordinator.
You asked
Answered from the demo venue guide and your booking only. Not from the open web.
Evening guests would go from 60 to 70. This costs money, so it needs a person to say yes.
Example price, not a quote. Nothing is booked until Hannah confirms.
From: venue guide, p.11 From: your booking, guests
The Glasshouse holds 120 in the evening, so 70 fits.
Needs your coordinator’s OK
Hannah will reply by the end of tomorrow, Tuesday 6 October. By email, to the address on your booking.
What Hannah receives
Booking attached: 80 day guests, 60 evening guests.
The reply time is an example, not a promise from any venue.
Hannah needs your final guest numbers by Monday 19 October. The kitchen orders from them.
From: your booking, guests
Hannah’s daily summary for Fernhill (demo), Monday 5 October. It updates as the couple taps. One summary of every question, so coordinators see it all at a glance.
My choices behind this sketch
Sketch: an invented demo venue with made-up prices and dates. Not a Gilchrist venue, price list or quote.
The problem Picture our wedding in a room we can actually book, and know roughly what it will cost, before we visit.
What I’d test first Do couples who see the price build as they choose send fewer changes after the viewing? Not tested yet.
Indicative total £0
Indicative only. Your coordinator confirms the real price and what is free on your date. All figures are examples.
Your picture and choices went as one quote request. Your coordinator checks the date, confirms the price and replies. Nothing is booked yet.
Your coordinator’s copy is shown beside the phone, or below it on a small screen. In this sketch nothing is really sent.
Sketch: in the venue, the same view would run through your phone camera. Not built.
Preview, not sent yet
Nothing reaches the team until the couple taps Send to my coordinator.
My choices behind this sketch
Sketch: a demo venue with invented rooms, rules and times. Not a real Gilchrist venue.
The problem Know the day will work before it is booked, and give the team accurate sheets drafted from the couple’s choices.
What I’d test first Do couples fix a flagged clash with one tap rather than writing to the coordinator? Not tested yet.
Starting points are filled in. Change anything and see the day rebuild.
Sketch: the couple adds 4 guests and tells you one guest has a nut allergy.
Three sheets written from the couple’s choices. They update as the couple changes them.
My choices behind this sketch
I begin by sitting with the people doing the job and watching how it really goes. Then I count where the time goes, pick the jobs that pay back most for least effort, and build the smallest useful version. The people who use it test it, and I measure the result: time saved, errors, revenue. If it can be counted, it gets a number.
Simplest answer first. Sometimes a better form or a change to the process beats AI, and then that is what I recommend.
Each one follows the same rule: the AI drafts and a person decides. None has a result measured in hours yet, so each one says what can be seen, not a made-up number.
All designed and built by me, using AI coding tools.
The lead example
What it would do here: sort new enquiries across the venues, draft follow-ups for coordinators to check and send, nudge suppliers who have gone quiet, and draft a weekly report for head office, so people have more time for couples.
Built with: Python, Claude, Google Workspace, Slack, HubSpot
Built with: Claude, Slack, Google Calendar, Cloudflare Pages
Built with: Next.js, FastAPI, Stripe, Claude
Built with: HTML, CSS, JavaScript, Python
Built with: Python, FastAPI, PostgreSQL
Built with: Python, Claude
One assistant the whole company could share. It would sit over the tools you already work in: your customer records (CRM), email and calendars, the documents in SharePoint or Google Drive, each venue’s website, the tour and booking tools, and finance.
Anyone asks it in plain words, and it answers from your own documents, showing where each answer came from. Each person sees only what their role allows. It drafts, and people decide.
Designed, not built yet. It is the team version of the assistant I use every day.
Anyone asks in plain words. It answers from your own guides and procedures and shows the source.
For exampleA seasonal waiter could ask: where do the ceremony chairs go at Highley?
Who decidesThe person asking. If it is not written down, it says so and points them to their coordinator.
Looks into people, companies, suppliers, competitors and venues that could join the group. Each becomes a short, dated briefing with its sources.
For exampleA sourced profile of a country house before the first call to its owner.
Who decidesWhoever makes the call.
Pulls scattered information onto one page.
For exampleEach person’s morning brief. A weekly roll-up of enquiries, tours and bookings by venue for the CEO. A monthly summary of what couples praise and complain about in reviews.
Who decidesThe reader, on what to act.
Writes replies, follow-ups, staff and kitchen sheets, and reports.
For exampleA reply to a couple’s enquiry, written from that venue’s prices and guide.
Who decidesA person checks and sends every one. Nothing sends itself.
A question becomes research, then a quick sketch, then a test with real couples or staff, then a decision. Ideas are checked again every week.
For exampleThe research lab in section 04 already works this way.
Who decidesYou, at the decision point.
One place for all of it, so what the business knows is not spread across inboxes, spreadsheets and people’s heads.
A worked example
Every Monday it lists the unsold dates in the next six months across all twelve venues. Most are winter and weekday dates, the ones your late-availability offers already aim at.
Who decidesSales, on which dates to work first.
For each date it finds who might want it: couples who enquired or toured but did not book, couples who asked for a date that was already taken, and what nearby venues charge for the same weeks.
Who decidesNothing goes out. It only gathers.
A personal offer for each couple, written from the venue’s own package and late-availability prices.
Who decidesThe sales consultant checks and sends. Nothing goes out on its own.
If one kind of date keeps going unsold, it suggests a new package to test, for example a smaller midweek wedding. It is tried with a few couples and the results come back before anyone commits.
Who decidesYou decide whether to test it, and whether to keep it.
The weekly roll-up shows which offers turned into tours and bookings, by venue, so the next Monday’s list gets better.
Who decidesHead office, on what to do more of.
Three layers
CEO
Head office (finance and HR)
Venue coordinator
Wedding-day staff
Seven steps from a couple’s first message to head office’s morning numbers, and who decides at each one.
Step 1 of 7
A couple’s enquiry arrives on a venue website.
Who decidesThe couple.
Step 2 of 7
It lands in your customer records with the venue and the date, and the assistant checks whether the date is free.
Who decidesNothing to decide yet. It only checks.
Step 3 of 7
It drafts a reply from that venue’s prices and guide, showing its sources.
Who decidesThe sales consultant edits it and sends it.
Step 4 of 7
A tour is booked, and the coordinator gets a short brief on the couple before they arrive.
Who decidesThe coordinator runs the tour.
Step 5 of 7
The booking is confirmed. The couple’s choices feed the staff, supplier and kitchen sheets, and allergies go to the head chef.
Who decidesThe head chef signs off allergens. The coordinator checks the sheets.
Step 6 of 7
On the day, staff ask the assistant on their phones, for example where the ceremony chairs go. Answers come from the venue’s own procedures.
Who decidesStaff. If it is not written down, it tells them to ask the coordinator.
Step 7 of 7
Next morning, head office sees enquiries, tours and bookings by venue in one roll-up.
Who decidesHead office, on what to act.
One change, every place. When a venue guide or menu changes in your documents, the website, the brochure and the assistant’s answers could all be updated from that one change.
Finance and HR stay with head office. A coordinator sees their own venue’s couples, never another venue’s. The limit is checked every time it looks something up, not just hidden on screen.
Each answer names the document and section it came from. If something is not written down, it says: I don’t know, ask your coordinator.
It drafts and a person sends. Every approval records who made it.
Allergies are health information, so they need the strictest handling, and a written plan would come before any pilot. A data-protection pack for this design is drafted, with a short risk assessment and a breach procedure, and would need a solicitor’s review.
Your stated aim is 25 venues, up from 12. Each new one could start like this.
Connect email and calendar
Planned at about 15 minutes.
It learns the venue
Short questions over the first week about how this venue runs.
Shared procedures from day one
Group-wide procedures apply as soon as the venue joins.
Counted from the day the build starts, for one venue. The first month in the job is always spent with your teams, so the build starts after that.
About three weeks
Staff at one venue can ask questions on their phones and get answers from that venue’s own guides, with the source shown.
About six to eight weeks
The full pilot at that venue: each role sees only its own part, enquiry replies are drafted for the sales team to send, and each coordinator gets a morning brief.
Then, venue by venue
A month of measuring, then the next venues, with what the first one taught us.
My estimate, not a promise. The biggest factor is how much of your know-how is already written down, and which systems hold your enquiries today.
First step: a pilot with one venue’s procedures and its coordinators, measuring how many questions it answers correctly and how often people still ask a person.
What the advert asks for, and where I have already done it. Where I have not, I say so.
My AI assistant now does my daily trawl through inbox, calendar, files and CRM, and I review what it finds. The prototypes for small firms start from one slow process.
My AI assistant connects Gmail, Calendar, Drive, Sheets, Slack and HubSpot, and runs on a schedule. In agency work I built steps in automated customer journeys, alongside the journey builders, and the personalised emails they sent.
Every system on this page is built in Python or JavaScript, with databases and outside services connected through their APIs.
I build with Claude and pick the approach by the job. The transit platform and the shop assistant each use AI in a different way.
Prototypes for small firms, shown live before anyone pays. My AI assistant has run daily since April 2026, and over 2,000 automated checks run against it.
Every AI I build drafts and a person decides. It cannot send anything on its own, keeps passwords out of what it writes, and has rules on what it may touch.
In agency work I trained and onboarded junior colleagues and wrote the process documentation.
Sometimes a better form or a change to the process beats AI. I say so first.
Each venue seems to run its own website, brochure form and booking calendar, so I would start by joining enquiries, tours and bookings into one view.
Local: I live in East Grinstead.
Contact: linardsberzins@gmail.com · 07949 783 799 · CV (PDF)