A hospitality group's loyalty programme is rarely killed by a bad platform choice. It's killed on a Tuesday in February, when the GM is short two waiters, the marketing coordinator is editing a Christmas menu six months early, and nobody pushes the campaign that was supposed to lift midweek covers. That's the moment that decides whether you should hire a loyalty agency or build it in-house: not a spreadsheet on cost, but an honest answer to who is going to run the thing when the week goes sideways.
- The real question isn't agency vs in-house. It's whether anyone owns the programme day to day.
- In-house only works once you have a dedicated CRM or loyalty role, not a marketing generalist with loyalty on a list of 14 things.
- Agencies are worth it when the bottleneck is creative cadence and segmentation, not platform choice.
- Measure either model the same way: 90-day repeat rate, member AOV vs non-member AOV, and 12-month LTV.
- Platform decisions take a day. Programme decisions take 12 months of consistent campaigns to compound.
The Real Question Isn't Agency vs In-House. It's Who's Running It on Tuesday.#
Most articles on this topic frame it as a binary on cost and control (LoyaltyPoint, Evolve My Media). That framing assumes the in-house option actually exists. In most multi-site hospitality groups under £15M in revenue, it doesn't. There is no "loyalty programme manager" in the org chart. There's a marketing director splitting attention across PR, menu launches, social, supplier shoots, and three openings in the pipeline. Loyalty is item 11 on a list of 14.
That's why most hospitality loyalty programmes look the same after 18 months: the software is paid for, a few thousand members are enrolled, and the last campaign push went out in October.
"We bought the platform in March. By September we had 4,000 members, no campaigns scheduled, and the GMs had stopped asking customers to join because nothing was happening with the data."
What Building It In-House Actually Costs (Beyond the Software Fee)#
If you're going to run loyalty in-house properly, here is the real cost stack for a hospitality group. The software line is the smallest one.
| Cost line | Realistic annual figure (UK, 5-15 sites) | Notes |
|---|---|---|
| Loyalty platform / wallet pass software | £3,000 to £12,000 | Stamp, points or membership card depending on model |
| Dedicated CRM / loyalty coordinator (0.5 to 1.0 FTE) | £18,000 to £42,000 | Salary + NI + pension. The role most groups skip. |
| Creative production (campaign artwork, copy, photography) | £8,000 to £20,000 | 12 to 24 campaigns per year if you push monthly |
| Reporting and analytics setup | £2,000 to £6,000 | Repeat rate, cohort retention, AOV by segment |
| Staff training and refresh cycles | £2,000 to £5,000 | Critical given hospitality's ~70% annual frontline turnover |
| Total realistic in-house cost | £33,000 to £85,000 / year | Versus the software fee alone, which most groups budget for |
The number that breaks most in-house plans is the coordinator. Without that role, the platform sits idle. With it, you've added a £30k+ salary to run something that may or may not lift repeat rate enough to justify it. That's the calculation. It's not about whether agencies are "better". It's about whether you'd genuinely hire and retain a CRM specialist for a 7-site group.
Five Signals a Loyalty Agency Is the Right Call#
- You've had a loyalty programme live for 6+ months and can't name the last campaign you pushed.
- Your marketing lead lists loyalty as one of more than five concurrent responsibilities.
- You can't currently report on member vs non-member AOV or 90-day repeat rate without manual export work.
- GMs have stopped actively promoting sign-ups because there's no visible activity in the programme.
- You're opening or have recently opened a new site, and the marketing team's bandwidth is fully consumed by launch work.
Three or more of those and the in-house model is already failing. The fix isn't a better platform. It's an operator.
When In-House Genuinely Makes Sense#
There's a threshold where in-house starts to win, and it's clearer than most agencies will admit:
- 15+ sites or £20M+ in revenue, where a dedicated CRM hire is justified by member volume alone.
- An existing data and analytics function (not just a marketing team) that owns reporting infrastructure.
- A loyalty strategy that depends on deep proprietary segmentation across multiple brand concepts under one group.
- Leadership that has committed to loyalty as a 24-month investment with quarterly LTV targets, not a tactical lever.
Below that threshold, an agency model almost always outperforms in-house simply because the work gets done. Consistency of execution beats sophistication of strategy every time, especially in hospitality where the loyalty mechanic itself (stamp, points, membership) isn't where the leverage sits.
What an Agency Should Own vs What Your Team Should Own#
| Owned by agency / managed service | Owned by your in-house team |
|---|---|
| Campaign calendar and creative production | Brand voice and creative approval |
| Wallet pass design and updates | Site-level rollout and GM briefings |
| Segmentation logic and reward economics modelling | Commercial sign-off on reward value |
| Reporting on repeat rate, AOV uplift, LTV by cohort | Reviewing reports and setting growth targets |
| Staff training materials and refreshers | Operational enforcement at site level |
| Platform configuration and ongoing optimisation | Integration with POS workflows where relevant |
How to Measure Whether Either Model Is Working#
Three metrics decide it, and they're the same whether you run loyalty in-house or hand it to an agency. If you can assess whether your current programme is actually driving revenue, you'll know within a quarter whether to change the model:
- 90-day repeat visit rate of enrolled members vs non-members. A working programme shows 15 to 35 percentage points of lift within 6 months.
- AOV delta between members and non-members on the same daypart. Healthy programmes deliver 8 to 18% higher member AOV.
- 12-month LTV by enrolment cohort. The single most important number, and the one most in-house programmes never measure because nobody owns the analysis.
If those numbers are flat after two quarters, the model is wrong, regardless of which side of the agency/in-house line you're on.
The Decision Framework#
| Your situation | Likely right model |
|---|---|
| 1 to 4 sites, founder-led, no dedicated marketer | Managed agency, light touch |
| 5 to 14 sites, 1 to 2 person marketing team | Managed agency or hybrid (agency runs, you approve) |
| 15+ sites, dedicated CRM hire on payroll | In-house with specialist agency support on creative and analytics |
| Pre-opening or in active expansion phase | Agency, regardless of size. Bandwidth is the binding constraint. |
| Loyalty already live but inactive for 3+ months | Agency. The in-house experiment has already produced its result. |
How long should we give an in-house loyalty programme before deciding to outsource?
Two quarters of measured activity. If you can't show campaign cadence (at least one push or update per month) and movement on member repeat rate or AOV after six months, the in-house model isn't working. Extending it rarely fixes it because the bandwidth constraint doesn't change.
Can we run loyalty in-house with just our existing marketing team?
Only if loyalty is genuinely their top-three responsibility and they have analytics support. In most hospitality groups under 15 sites, the marketing team's calendar is dominated by openings, PR, social and menu work. Loyalty becomes the thing that slips.
What does a managed loyalty programme typically cost vs an in-house hire?
A managed service for a 5 to 15 site group usually lands between £15,000 and £40,000 per year all-in. A dedicated CRM hire with the same scope (salary, oncosts, tools, training) lands between £40,000 and £60,000. The gap closes above 15 sites. You can see <a href="/blog//pricing">what a managed loyalty programme costs vs an in-house hire</a> in more detail.
Is choosing a loyalty platform the same decision as choosing how to run loyalty?
No, and conflating them is the most common mistake. Picking a platform is a one-day decision. Building a programme that compounds repeat revenue is a 12-month commitment requiring monthly creative, segmentation work, and reporting. The platform doesn't do any of that on its own.
What if we already have software but no one's running it?
That's the most common starting point. Audit current member behaviour first (repeat rate, AOV, last campaign date), then decide whether to hire, outsource, or in some cases sunset the programme entirely and restart with clearer ownership.