Migrate from Xibo without losing
campaigns.

You're here because something isn't quite adding up. You've been running Xibo, Anthias, Concerto, info-beamer or whatever for months, and the phrase "it's free" is starting to sound like a punchline. The fourth Friday night you found yourself SSH-ing into a VPS because the menu screen went black, you added up your time for the first time. The number looked wrong.

This guide won't try to talk you into anything you don't want. It'll lay out the next steps without preaching. If you decide to stick with self-hosted in the end, we'll applaud you: there are five profiles where it genuinely is the right answer, and we name them without any sleight of hand. If you decide to try SaaS, we'll show you how to migrate without losing what you've already built.

Radical honesty: we're Digisini, a SaaS. We have an obvious interest in you trying our platform. But we wrote this guide the way we'd have wanted to read it when it was our turn. No bashing Xibo — it's good software for the people who need it. No hiding numbers. No promising things we can't back up.

The structure is linear. Audit first (what you have), TCO next (what it's really costing you), decision in the middle (go or stay), migration step by step if you go, and common mistakes so you don't have to make them yourself. Thirty minutes to read, two afternoons of work if you decide to act.

Let's go.

Before touching anything, know what you have. The main reason migrations go wrong isn't technical complexity — it's discovering on the day you switch off Xibo that there was a special events calendar scheduled for Saturday that nobody had written down. Thirty minutes of inventory saves a Friday night at eleven.

1.1 · Screen inventory

Open a blank spreadsheet and fill in one row per screen:

• Physical location. "Bar counter, Brighton branch", not "screen 3".
• TV and player model. Smart TV (Samsung, LG, Sony), Android TV box, Chromecast, Fire Stick, Raspberry Pi with SD card model X.
• Resolution and orientation. 1080p landscape, 4K vertical portrait.
• Network connection. WiFi, Ethernet, static IP or DHCP.
• Last time it failed and why. If you remember. If not, log it as missing data.
• Who last turned it on. You, the duty manager, an external engineer.

If you have more than five screens and this inventory doesn't already exist, leave the migration for tomorrow and just do this today. It's useful whether you stay on Xibo or move to SaaS. Nothing wasted.

1.2 · Content inventory

For each screen, list what's happening on it:

• Active playlists. Internal name, what they show, how often they change (daily, weekly, monthly).
• Special events calendar for the year. Christmas, Valentine's Day, Mother's Day, Halloween, Black Friday, football finals. This is the one people forget most.
• Compliance requirements. Allergens (the 14 under EU Regulation 1169/2011, Natasha's Law for pre-packed-for-direct-sale food in the UK), over-18 notices if you sell alcohol or vapes, legal capacity notices if you're a nightclub, licence number if you run a B&B or vacation rental.
• Organised media folder. By vertical, by expiry date, by season. If your folder is currently library/ with no structure, note it down: you'll need to sort it out.

1.3 · Operational inventory

The uncomfortable questions — but more important than the technical ones:

• Who manages this today? You, a manager, a cousin, an external agency, nobody.
• How much time do they spend on it per month? If you don't know, estimate high: most people underestimate by 2-3×.
• How many technical incidents in the past 6 months? Only the ones that reached you — the ones that didn't also count (someone had to turn the TV off and on again).
• What would happen tomorrow if that person left? If the answer is "it goes dark", you have a more urgent problem than TCO.

1.4 · Dependency inventory

Everything plugged into Xibo that you sometimes forget about:

• Integration with a till or POS (Square, Shopify POS, etc.)
• Booking software (OpenTable, Resy, Mews, Cloudbeds, Lodgify)
• Queue management with a physical ticket dispenser (Qmatic, Sipac, homebrew)
• Custom modules a developer wrote two years ago
• Cron scripts that pull media from an FTP
• Webhooks to Slack or a phone number that alert when something goes down

Each of these integrations is a decision you need to make before migrating. Most can be recreated on any modern SaaS with a bit of patience. Some (custom modules) need rewriting or compromise. You identify them now so there are no surprises halfway through the switch.

TCO stands for Total Cost of Ownership. For open source digital signage, almost nobody calculates it correctly the first time. The mental trap is counting only the visible costs (VPS, domain) and forgetting the invisible ones (your time). Here's how we unpick that trap.

2.1 · Use the Excel calculator attached

We've attached an Excel with six tabs:

1. Read me first — how to use it in five minutes.
2. Your current situation — six editable inputs.
3. Your real 3-year TCO on open source — generated by formulas.
4. Equivalent cost on Digisini — all five plans with their numbers.
5. Your net saving — the conclusion, in pounds and in hours.
6. Tweak the assumptions — so you can test whether the answer still holds with your own numbers.

Fill in the inputs and let the sheet do the maths. Five minutes.

2.2 · Why your first instinct is wrong

If you've been running Xibo in production, you probably think it costs you about £10-15/month (the VPS) plus "some time I can't really quantify". The real figure, for a business with 2-5 screens, is usually 2-3× that. Let's break it down.

The VPS is not the whole infrastructure cost

The VPS is £10/month. But you also need: a domain (£12/year), an SSL certificate (free if you use Let's Encrypt), external backups to a second provider so you have something when the main VPS burns down (Hetzner Storage Box is about £30/year for 1TB), and basic monitoring so you find out when something goes down (Uptime Robot free for one basic check, £50/year if you want SMS alerts).

Sysadmin hours count, whoever they come from

If you pay an external sysadmin, count their hours at their real rate. If you do it yourself, count your hours at the rate you charge clients when you invoice. If a family member does it for free, count the hours anyway (their time has value, even if they don't bill you): if they get tired of it tomorrow, you'll need to hire someone.

The most painful thing to account for is the owner's time. When the venue closes and you stay another hour on the laptop because "the menu screen is playing up again", that hour counts. Multiply by the real number of incidents per year and a big number starts to emerge.

Downtime has an opportunity cost

When the menu screen is dark on a Saturday afternoon, you don't lose immediate sales (people still walk into the café), but you lose signal, professionalism, and eventually a customer who walks in, sees a printed sign stuck up with tape, and heads next door instead. It's hard to measure, but pretending it doesn't exist is self-deception. The calculator models this as one 4-hour event per year at your hourly rate. If that seems conservative, raise it.

2.3 · Three real cases with numbers

Let's put numbers to three typical business sizes. Assume an external sysadmin at £30/h and an owner at £25/h.

Case A Café with 2 screens

Case B Small chain with 5 screens

Case C Network of 30 screens multi-site

At this size the maths complicates in favour of self-hosted — but only if you have in-house IT. Assume a part-time dedicated sysadmin (15% of an FTE) plus redundant infrastructure: ~£9,000/year, compared to £1,990/year for Digisini Network. Cloud still wins, but the £7,010/year gap needs to be divided by the reality that the sysadmin probably does a lot more beyond signage.

Once you're at 80-100+ screens with a full-time in-house sysadmin, self-hosted starts to break even or pull ahead on cost. But at that point you're not an SMB anymore — you're a mid-size retail operation, and this guide probably isn't aimed at you.

This section exists out of honesty. Not all situations are the same, and open source self-hosted is still the right choice for five specific profiles. If you recognise yourself in any of them, read the rest of this guide as background, not as a push to migrate.

3.1 · You have in-house IT and at least 30-50 screens

The sysadmin cost is already covered by other work: they maintain the ERP, the CRM, the file servers. One more Xibo instance is marginal. And the freedom to modify code is genuine, not decorative: it lets you tune dayparting to the minute, integrate with your internal queue management system, write a module that shows the Power BI dashboard on the boardroom screen.

3.2 · Non-negotiable data sovereignty

Banking, regulated defence, healthcare with data in specific countries, public sector with "data doesn't leave here" clauses in their contracts. There's certified cloud for many of these cases, but sometimes it's faster and cheaper to host it yourself in a rack you own at the datacentre that came with your tender. If this is your reality, Xibo or Concerto on your own infrastructure are still reasonable choices.

3.3 · You need to extend the code

Your business depends on a module no SaaS will build for you. An exotic SAP integration, RFID readers in kiosks, syncing with the shopping centre's internal radio, a connector to the PMS you migrated from 15 years ago that nobody wants to touch. Forking an open source project, spending a couple of weeks on development and maintaining it is reasonable. A SaaS will tell you no, or quote you Enterprise.

3.4 · You're an agency or integrator selling the service

You sell clients a white-label solution with Xibo or info-beamer underneath. It's your business model. You take maintenance seriously because you're paid to. There's a whole world of certified Xibo partners making a very good living from arrangements like this.

3.5 · You're a maker and learning is the reward

A Raspberry Pi running Anthias in your workshop, at home, or in your coworking space. It's fun, you learn, you're not monetising it, you don't mind if it drops out on Sunday afternoon. We're not here to judge the fun. But let's be clear: it's not comparable to a business that needs the menu screen alive on a Saturday at two.

We're assuming you've done the audit from Chapter 1, calculated the TCO from Chapter 2, and decided that SaaS is the right move. Now we execute. The target: one afternoon for a 2-3 screen business, two afternoons for 5-10 screens, one relaxed week for 20+.

Step 1 · Export media from Xibo

The most voluminous part, but the most straightforward. You have two routes:

• Via SFTP to the VPS. Download the whole /var/www/.../library/ folder. Use FileZilla, Cyberduck, or scp from the terminal. Typical sizes: 2-10 GB.
• Via the Xibo UI. Library → select all → Download. Slower, but doesn't require SSH access.

While the download runs (it's the slow part), do a clean-up: delete last year's videos you're no longer using. Don't upload them to the SaaS "just in case". That "just in case" inflates your storage quota for no reason.

Step 2 · Export layouts and schedule

Xibo lets you export each layout as a .zip file: Layouts → Export. It's not an industry standard, so the SaaS you're moving to won't import it directly. But those zips are useful as visual reference: unzip them, look at the XML, see which media went in which zone, what duration order items had, what dayparting rules they carried.

More practical than the zip: take screenshots of your Xibo calendar in the monthly and annual view. Screenshot each active layout in the editor. Think of it as taking a mental print: you're going to rebuild it, not copy it.

Step 3 · Functional equivalences Xibo → Digisini

Step 4 · Sign up to Digisini and connect the first screen

It's quick. Five minutes:

1. Go to app.digisini.com/register. No card needed.
2. Choose the Free plan to start (you can always upgrade). 1 screen, with a watermark.
3. Pick your vertical (café, gym, dental, driving school, etc.) and load the template.
4. Swap in your logo, brand colours and contact details from the dashboard.
5. On the physical TV, open signage.digisini.com in the browser — it shows a 6-digit PIN. Type that PIN into your dashboard from your phone, and that screen is registered.

Step 5 · Rebuilding playlists

The most manual part. For each playlist you had in Xibo, you do the same in Digisini: drag media into the order you want, set durations, mark zones if the screen has a split layout. The key difference: in Digisini you start from vertical templates with menus, allergens, timetables or queue management already structured. That saves hours of design from scratch.

If you had 6 playlists in Xibo, expect around 45 minutes per playlist. The first two take longer (learning curve with the editor), the last few fly.

Step 6 · Time scheduling (dayparting)

This is where you load the calendar you documented in Chapter 1. Mother's Day event, Valentine's Day, Christmas, football finals. Scheduling by time slot (lunch menu vs dinner menu). Weekend-specific promos.

Tip: schedule all the fixed events for the whole year first. Variable-date ones you configure when they approach. This removes the nagging anxiety of "I'm going to forget Halloween" — because it's already in there.

Step 7 · Multi-site (if applicable)

If you have more than one venue, don't migrate them all at the same time. Do one as a pilot for a week. If that week reveals no unexpected issues, roll it out to the rest. Digisini's multi-site centralisation from the Pro plan lets you manage everything from one dashboard: update the menu for the Brighton branch from your sofa at home, and it propagates to the right screens.

Five traps that come up again and again. We've seen them enough times to name them.

5.1 · Forgetting the year's special events calendar

The king of mistakes. You migrate in April, everything works, you get to October and realise that Halloween, Christmas, Valentine's Day and Black Friday aren't scheduled. Document everything before you switch off Xibo, and load all twelve months into the SaaS calendar even if you're only in April.

5.2 · Not documenting the schedule before shutting Xibo down

Once you kill the VPS and cancel the hosting subscription, recovering the schedule is very hard. Before you switch off: take screenshots of the monthly calendar, export layouts as zips, save a copy of the Xibo database SQL dump to your local drive (not the VPS you're about to delete). Thirty minutes of insurance.

5.3 · Trying to migrate everything in one day

If you have 12 screens, don't move them all on the same Friday. Migrate one as a pilot, give it a week, then roll out to the rest when you're relaxed. The pilot reveals the nuances the manual doesn't cover. A mass migration reveals all of them at once, in the middle of your busiest weekend.

5.4 · Ignoring allergens with "we'll sort that later"

If you're in hospitality and the 14 allergens required under EU Regulation 1169/2011 (and Natasha's Law in the UK for pre-packed-for-direct-sale items) aren't on screen from minute one, you're exposed to a regulatory sanction. It's not a cosmetic detail. Digisini's vertical templates include allergen fields already structured — you just fill them in.

5.5 · Not warning the venue manager that the panel is changing

The venue manager has muscle memory with the old dashboard. If you turn up on a Monday having switched the tool without telling anyone, they'll resent the migration. Fifteen minutes of "here's what's different, here's what's the same" saves friction and operational errors down the line.

There will be a moment in the migration where something doesn't go as expected. Three pieces of advice to stop that moment turning into abandonment.

6.1 · When to escalate to support

If you've been stuck on the same step for more than 30 minutes and you're not moving forward, escalate. It's not weakness — it's efficiency. Support will resolve in 15 minutes what it would take you 2 hours to figure out from documentation.

6.2 · How to write a useful support email

Include:

• What you were trying to do ("I'm setting up a playlist for the lounge screen")
• What you literally did ("I clicked Templates → Café → Edit")
• What you expected to happen ("the template to load")
• What actually happened ("got a 500 error")
• Screenshot of the error
• Exact URL of the dashboard page you were on

This saves 3 rounds of "could you send me a screenshot?". Support replies in hours rather than days.

6.3 · What to expect on response time

Plan Free and Starter: email, reply within 24-48 working hours. Plan Pro and above: priority support, average response under 4 working hours. Network and Enterprise: dedicated support.

If your case is urgent (screens down during peak hours) and you're on Free/Starter, the best trick is to sign up for the Pro trial (14 days, no card), which puts you in the priority queue. On day 15 you decide whether to stay. We say this with no strings attached.

We answer in English (and Spanish if you prefer).