Thai, Oil or Deep Tissue: Which Massage to Book in Bangkok

Every massage menu in Bangkok lists the same three names, and almost none of them explain the difference. Every week we publish one practical thing on recovery: what the evidence supports, what it does not, and how to spend your hour well. Stay in the loop.

What Actually Separates Thai, Oil and Deep Tissue?

Short answer: Traditional Thai massage is assisted stretching and rhythmic compression, done fully clothed on a mat with no oil. Oil massage is continuous gliding work with warm oil on a table. Deep tissue is slow, targeted pressure aimed at a specific layer of muscle. Three different experiences, three different reasons to book.

Start with the mechanics, because the marketing language is useless here.

Traditional Thai massage, or Nuad Thai, is the one that looks least like what most people picture. You keep your clothes on. There is no oil and no table. You lie on a mat while the therapist works with palms, thumbs, forearms, elbows, knees and feet, combining steady compression along the body's energy lines with assisted stretches: hips opened, spine rotated, shoulders taken through range.

It is a recognised tradition rather than a spa invention. UNESCO inscribed Nuad Thai on the Representative List of the Intangible Cultural Heritage of Humanity in 2019. The listing calls it a non-medicinal remedy and manual therapy that "involves bodily manipulation in which the practitioner helps rebalance the patient's body, energy and structure."

The World Health Organization published a training benchmark for it in 2010, and that document makes a distinction the menus rarely do: Nuad Thai splits into two types, which the document calls Nuad Thai for health and Nuad Thai therapy. The first, in WHO's own words, is "intended for relaxation and health promotion rather than for therapeutic purposes". The second is targeted work on a complaint. Same tradition, two different jobs.

Oil massage covers the aroma oil and Swedish family: long, continuous strokes with warm oil, on a table, under a towel. Contact stays unbroken, pressure sits medium at most, and the pace is deliberately unhurried. Worth knowing, because it gets dismissed as the unserious option: oil work is part of the formal Thai professional curriculum, not an alternative to it.

Deep tissue is the most misunderstood of the three, and the confusion starts with the name. Plenty of clinical descriptions define it by force, as slower and firmer strokes reaching deeper layers of muscle and connective tissue. That is a fair description of what it feels like, but it makes force sound like the active ingredient. A 2018 review in the Journal of Bodywork and Movement Therapies looked at how the term is actually used and found it inconsistently defined across the literature, with the manual techniques it names, friction across the grain of the muscle and stripping along it, doing more to describe the work than pressure alone. What separates it in practice is a slow pace aimed at a particular layer, held long enough for tissue to release instead of brace.

That is the honest taxonomy. Clothed and stretched, oiled and glided, or slow and targeted. Everything else on the menu is a variation on those three.

Which One Should You Book? Start With How You Feel

Short answer: Book by state, not by name. Stiff and short in range points to Thai. One specific area that has hurt for weeks points to deep tissue. Wound up, sleeping badly, running on empty points to oil work. If you cannot tell, describe the feeling when you book and let the therapist choose.

Most people pick a massage the way they pick from a wine list they do not understand: by price, or by whichever name sounds most serious. That is how a person with three weeks of bad sleep ends up booking deep tissue, and leaves more wired than they arrived.

A better filter takes one question. Not "what do I want to buy", but "how does this actually feel".

The distinction between the last two is worth spelling out, because it is where most bookings go wrong. Deep tissue asks a question of one area and waits for an answer. Oil work does the opposite: it keeps the hands moving, keeps contact unbroken, and lets the whole body settle. One is investigation. The other is a slow landing.

There is a version of this question that people are embarrassed to ask, so here it is plainly: it is completely normal not to know. Describing the feeling is enough. Any therapist worth the hour would rather hear "my shoulder has been complaining for a month" than "deep tissue, please."

Is Deep Tissue Actually Stronger Than the Others?

Short answer: Stronger, yes. More effective, no. When researchers compared massage types head to head, they found no meaningful difference in results. A trial of 401 people with chronic low back pain found relaxation massage and structural massage worked equally well, and the relaxation group held its gains slightly longer.

This is the part that upends the usual advice, so it is worth going slowly.

The intuition is that harder work does more. Pay more, press harder, get better results. Three separate trials say otherwise.

The largest is Cherkin and colleagues, published in Annals of Internal Medicine in 2011, with 401 participants who had chronic low back pain. They compared structural massage, the technical, problem-solving kind, against straightforward relaxation massage. The finding:

"No clinically meaningful difference between relaxation and structural massage was observed in terms of relieving disability or symptoms." - Cherkin DC et al., A Comparison of the Effects of 2 Types of Massage and Usual Care on Chronic Low Back Pain, Annals of Internal Medicine, 2011

The authors also noted that the benefit of relaxation massage on function, though not on symptom severity, was still measurable at 52 weeks, and small.

The same pattern shows up when Thai massage goes head to head with Western technique. A Khon Kaen University trial of 180 patients with back pain linked to myofascial trigger points compared traditional Thai massage with Swedish massage over six sessions. Both groups improved significantly. Neither group beat the other.

A third comparison, smaller and against a physiotherapy technique rather than another massage, points the same way. Forty-five patients, Thai massage against muscle energy technique:

"Both TM and the ME technique resulted in a significant improvement in all parameters (p < 0.05) compared to the control group. Additionally, no significant difference was observed between TM and the ME technique in all parameters." - Buttagat V et al., Effects of traditional Thai massage versus muscle energy technique, Journal of Bodywork and Movement Therapies, 2021

Two head-to-head massage comparisons and one against a physiotherapy method, and not one of them found a winner. Nobody has yet demonstrated that the harder or more technical modality does more.

Which reframes the whole question. Choosing between Thai, oil and deep tissue is not choosing between weak and strong medicine. It is choosing an experience you can tolerate and will come back to. That matters more than it sounds, and the next sections explain why.

If the technique is not what separates a good hour from a wasted one, something else has to be. What separates them is whether the sessions add up to anything, and that is a question of programme rather than booking.

Is Thai Massage Supposed to Hurt?

Short answer: No. Firm is the goal, painful is not. The WHO training standard lists excessive pressure under adverse reactions, not under signs of good work. If you are holding your breath or bracing, the pressure has passed the point where it helps and started working against the session.

This is the single most common question people ask about Thai massage, in several forms: is it supposed to hurt, should it hurt to work, why does it hurt so much. The frequency of the question tells you how widespread the belief is.

The belief is wrong, and the official documentation says so plainly. The WHO benchmark document for Nuad Thai training handles excessive pressure under the heading of adverse reactions. Pressure that causes pain is listed as a thing to avoid and manage, not a thing to aim for.

There is a mechanical reason as well as an official one. A muscle that feels threatened contracts. If the pressure makes you tense up and hold your breath, the tissue underneath is now bracing against the very hands trying to release it. The session starts working against itself. Firm pressure you can breathe through is the target; anything past that is subtraction, not addition.

The practical version: you are allowed to say so. Not at the end, in the feedback form. During, at the moment it happens. "That is too much" is a complete sentence, and a good therapist will thank you for it, because they have been guessing until you said it.

Worth adding for anyone whose only reference point is Thai massage on a beach: the tradition itself does not equate depth with quality. The WHO split between practice for health promotion and practice for therapy exists precisely because one tradition covers both a calm restorative hour and targeted clinical work.

So What Does Decide Whether It Works?

Short answer: Dose. In the only trial designed specifically to test it, 30-minute sessions produced no better results than being on a waiting list. Sixty-minute sessions, two or three times a week, produced clear improvement. Session length and frequency moved the outcome. Technique did not.

Here is the finding that should change how you book, and that almost no massage menu in Bangkok mentions.

Sherman and colleagues ran a dosing trial for chronic neck pain, published in Annals of Family Medicine in 2014, with 228 participants split across six groups over four weeks. Some received 30-minute sessions two or three times a week. Some received 60-minute sessions once, twice or three times a week. One group was a waiting list.

The 30-minute groups did not separate from the waiting list. Not on pain, not on function. Half an hour of massage, two or three times a week, for a month, produced results indistinguishable from no massage at all.

The 60-minute sessions separated, but only above a certain frequency. Twice and three times a week produced clear improvement in both function and pain. Once a week did not, which makes frequency and length a package rather than a choice between them:

"After 4 weeks of treatment, we found multiple 60-minute massages per week more effective than fewer or shorter sessions for individuals with chronic neck pain." - Sherman KJ et al., Five-Week Outcomes From a Dosing Trial of Therapeutic Massage for Chronic Neck Pain, Annals of Family Medicine, 2014

A follow-up study found that a later booster course of weekly 60-minute sessions helped at 12 weeks, but only for people whose original sessions had been the 60-minute kind.

Put that next to the comparison trials from the previous section and a shape emerges. Which technique you choose: no measurable difference. How long the session runs and how often you come: measurable difference, repeatedly.

This is why the decision that looks like a saving is usually the costly one. A 30-minute session is not a smaller version of the treatment. On the evidence available, it is a different thing entirely, and that thing did not work.

How Often Should You Actually Go?

Short answer: More often than most people book, for a shorter stretch than they fear. The trials that produced results clustered sessions: roughly six to eight of them, spread across two to four weeks, at 60 minutes or longer. After that, spacing out to maintenance is reasonable. One session on its own is a test, not a treatment.

The dosing evidence points at a shape rather than a rule. Cochrane's 2024 review of massage for neck pain ran a subgroup analysis by dose and found a clinically important difference favouring courses of at least eight sessions over four weeks. Its lower bound for session length is 30 minutes, which sits awkwardly next to the Sherman trial, where 30-minute sessions failed outright. The two are not really in conflict: Cochrane pooled a mixed set of studies and drew a floor, while Sherman compared lengths directly and found the floor too low. When one study is designed to answer exactly this question and the pooled set is not, weight the one that was. Book the hour.

For scale on the rest: the Thai massage trial mentioned earlier used eight sessions over two weeks, the Thai versus Swedish comparison six over three to four weeks.

So: a front-loaded block, then space it out. Not one session when it gets unbearable, then nothing for three months.

If your reason for booking is desk-related neck and shoulder pain specifically, that pattern has its own logic, and we wrote it up separately in our guide to office syndrome in Bangkok. Short version: the desk is still there on Monday, so frequency matters more than intensity.

One caveat on our own practice, stated plainly. We build individual programmes around repeated sessions, and after your first visit we write down what was worked on and how you responded, so the next therapist starts where the last one finished. That is our method, and it is why over three or four visits this stops being a series of massages and starts being a plan. It is not a claim that our particular schedule has been tested in a trial. The trials support the shape: repeated, full-length sessions, clustered rather than scattered.

Why Am I Sore Two Days Later?

Short answer: Usually because the work was deep and your tissue was not used to it, which is normal and passes within a day or two. It is not evidence that the therapist did something wrong, and it is not evidence they did something right either. Sharp pain, lasting pain, or bruising is a different matter and worth reporting.

Delayed soreness after a first deep session is common enough to be unremarkable. The sensation is closer to the day after a first yoga class than to injury: diffuse, achy, fading.

What it is not is a scorecard. There is a widespread belief that soreness proves the massage worked, and the comparison trials make that difficult to defend, given that gentler modalities produced the same results as harder ones. Soreness is a side effect of unfamiliar load, not a measure of benefit.

The line worth watching is between ache and alarm. Diffuse soreness that eases over a day or two sits inside normal. Sharp pain during the session, pain that gets worse rather than better over the following days, significant bruising, or numbness and tingling do not. Those are worth a message to the studio and, depending on the picture, a doctor.

If soreness is a concern for you, say so at the start rather than the end. Pressure is adjustable, and a therapist told "I bruise easily and I have work tomorrow" will run the session differently from one told nothing.

What the Evidence Supports, and What It Does Not

Short answer: The strongest evidence for massage is not about pain at all. It is about anxiety, mood and post-exercise soreness. For back and neck pain the effect is real but short-lived and rated low certainty. And two of the most repeated claims in the industry, flushing toxins and lowering cortisol, do not survive contact with the primary research.

This is the section that will cost us a booking or two, and it is the reason to write the article at all.

Where the evidence is strongest. A meta-analysis by Moyer and colleagues in Psychological Bulletin found the largest effects of massage therapy were on trait anxiety, the settled kind rather than the mood of the moment, and on depressive symptoms, with reductions the authors compared in magnitude to a course of psychotherapy. Separately, a 2017 review in Frontiers in Physiology concluded that massage after strenuous exercise could be effective for relieving delayed onset muscle soreness, with the effect strongest around 48 hours. If you book for stress or for training recovery, you are on the firmest ground the field has.

Where it is real but modest. For low back and neck pain, Cochrane reviews find benefit, but describe the certainty of the evidence as low and the effects as mostly short-term. That is not nothing. It is also not a cure, and nobody honest will sell it as one.

Where the popular story falls apart. Two claims deserve retiring:

• Massage flushes out lactic acid and toxins. A 2010 study in Medicine and Science in Sports and Exercise measured it directly and found massage after exercise actually reduced blood flow and the clearance of lactate compared with simply resting. It was a small study, twelve people and an isometric forearm task, so read it as a mechanism check rather than the last word. But the mechanism it found runs the opposite way to the story everyone tells.
• Massage lowers your cortisol. A meta-analysis by Moyer and colleagues in 2011 examined this specifically and concluded the effect on cortisol was indistinguishable from zero.

Both claims are everywhere in spa marketing, including from places that should know better. Neither is supported.

None of this makes massage less worth booking. It makes the reasons for booking more accurate, which is the only kind of reason worth having.

When You Should Not Book Any of the Three

Short answer: Skip all three and speak to a doctor first if you have a fever, an active infection, a recent injury with swelling, or a suspected or diagnosed blood clot. If you are pregnant or taking blood thinners, massage is not ruled out, but the studio has to know before you book. These are not formalities. Documented cases of serious harm exist, and every one of them turns on information the therapist did not have.

The WHO training benchmark for Nuad Thai carries a formal contraindication list. In plain terms, the situations that mean not today:

• Fever, or any active infection.
• A recent injury with swelling or bruising, or a fracture.
• A suspected blood clot, or a diagnosed deep vein thrombosis.
• Pregnancy, which does not rule out massage but does change what is appropriate and requires a therapist trained for it.
• Blood thinners, which do not rule it out either, but must be mentioned so pressure can be adjusted.

The reason to take this seriously rather than treat it as small print is that the case reports exist, and they are specific.

A 2020 report in BMC Pregnancy and Childbirth documented a life-threatening pulmonary embolism during leg massage at a commercial shop in Thailand. The patient was 25 weeks pregnant with a deep vein thrombosis nobody had diagnosed. She collapsed within minutes and was resuscitated; two months later, when the report was written, she remained unconscious with severe hypoxic brain injury, and the pregnancy had been lost. A separate report described rhabdomyolysis and acute kidney failure after intense massage.

The third case breaks the pattern, and that is why it matters most here. A 2026 report in Frontiers in Nephrology described renal infarction hours after a deep tissue session from an unlicensed practitioner. The patient was a healthy 50-year-old man with no underlying condition at all. The authors' point is that direct mechanical force can do damage on its own, without a hidden diagnosis to explain it.

So: two of the three turn on something nobody knew, and the third on pressure alone. Between them they cover both reasons to tell the studio what is going on and to stop the session when it stops feeling firm and starts feeling wrong.

Separately, there are symptoms where the answer is not a different massage but a doctor, promptly. The NHS lists these red flags for back pain:

Go to emergency care, not to a studio:

• Numbness, tingling or weakness in both legs.
• Loss of feeling around your genitals or anus.
• Difficulty passing urine, or loss of bladder or bowel control.
• Chest pain.
• Back pain after a serious accident, such as a car crash.

See a doctor before booking anything:

• Feeling hot, cold, shivery or generally unwell alongside the pain.
• Unexplained weight loss.
• A swelling or a change in the shape of your back.
• Pain that is no better after resting, or that is worse at night.

None of those is a booking. That is an appointment or, for the first group, immediate care.

How to Ask for What You Want

Short answer: Describe the feeling and the context, not the product name. Say where it hurts and for how long, mention anything medical, and name the pressure you actually want while the session is running. The difference between a good hour and a wasted one is usually information the therapist never received.

Five things worth saying out loud, in roughly this order.

1. What is going on, in plain language. "My left shoulder has been complaining for a month, worse in the afternoon" gives a therapist more to work with than any product name on the menu.
2. How long you have. A full-length session is not an upsell here, it is the variable with the evidence behind it. If your budget allows one properly long session a week rather than three short ones, the research suggests the long one.
3. Anything medical. Pregnancy, blood thinners, recent surgery, varicose veins, a fresh injury. This is the sentence the two serious case reports above were missing.
4. Pressure, during, not after. Firm should be something you can breathe through. If you cannot, say so at the moment, not in a review.
5. Whether you want to talk. Some people decompress by talking, some by not. Neither is wrong, and saying which saves both of you the guessing.

One more thing, which is less advice than an observation. The studios worth returning to are the ones that ask you these questions before you have to volunteer the answers, and that remember them next time. Consistency is scarcer than skill, and it is the part you cannot judge from a price list.

If you are choosing where to go in our part of the city, we wrote a practical guide to booking a massage in Ekkamai, including how to get there and what people in the neighbourhood tend to book.