I keep seeing this pattern in agency books and it drives me nuts. Worked years on the consulting side, now I do independent pricing work, and it's the same story every time.

Retainer margin sits at 18-22%. Owner sees the numbers, panics, decides the fix is to increase rates. Six months later they've lost 30% of clients and the rest are complaining about every change order. Rinse and repeat.

But that diagnosis is nearly always wrong. Below 25% retainer margin is rarely a pricing problem - it's a discovery problem.

Here's the actual maths. Retainers price for an assumed scope. Discovery defines that scope. If discovery is rushed (and it almost always is for repeat clients - "we know each other, we'll figure it out as we go"), the actual work drifts 20-40% above what was priced within 90 days.

Real example I looked at last year: a web dev agency in DACH. Their retainer was 40 hours per month at €4500, priced at 75% utilisation, so 30 billable hours expected. Month 3 actual delivery? 47 hours. Of those 47, only 32 were in the SOW. The other 15 were small stuff - quick fixes, "while you're in there, can you also do X". Margin collapsed from a projected 28% to an actual 11%.

The owner saw 11% and said, "we need higher rates." But raising rates wouldn't have fixed it - those 15 hours of out-of-scope work would still happen. Discovery is where you actually fix this.

So what works in practice?

Tighter discovery upfront, even for retainer renewals. A 90-minute scope review call once a quarter. Document current delivery against the original SOW. Drift is usually obvious once you write it down. Most agencies skip this because they trust the relationship.
Change order discipline. Anything outside documented scope is a paid change order. Agencies underestimate how much margin they give away by saying yes to small requests. 15 hours of small stuff per month is two days of unpaid work, every month.
Discovery as its own paid product. The real upsell sits before the retainer. Sell paid discovery at €3-5k. It qualifies the buyer (anyone willing to pay for discovery is serious) and surfaces scope before the retainer starts.

Raising prices on a broken pricing model just compresses retention. The retainer maths itself is usually fine - what needs fixing happens after signing.

If your margin sits below 25%, run this audit before you adjust anything. Take your last five retainer projects, list everything you actually delivered against everything in the original SOW. The gap is the real reason your margin is dying.