Seven steps to compare connected TV advertising reviews

What to take away

Why most connected TV advertising reviews mislead buyers

Connected TV buying is not one product but a set of overlapping offers: managed service, self-serve platform, broadcaster catch-up, publisher inventory, and programmatic access to streaming apps.

A review of one is not a review of another. A self-serve dashboard review covers an interface; a managed service review covers a relationship. Buyers often need one and read the other.

The common mistake is to read a score out of ten, decide it is good, and brief an agency on that basis. The score was produced against a brief that may have nothing to do with yours. Different budget, different sector, different creative, different measurement expectation.

Reviews also age quickly. Inventory deals change, measurement partners change, and pricing bands move. A review written eighteen months ago may describe a product that no longer exists in that form.

There is a commercial layer too. Many comparison pages are affiliate-funded, meaning the publisher earns a fee when a reader signs up. That does not make the facts wrong, but it does shape which products appear and how prominently. A ranking with ten entries and one obvious winner usually reflects a commission structure as much as a test.

None of this means reviews are useless. It means they are evidence, not answers. The job is to work out what a review actually tested, then decide how much weight it deserves in your own decision.

Step 1: write your brief before you read a single review

Write down what you are buying. Audience, budget band, campaign length, creative format, and the number you will judge success on. Keep it to one page.

This takes an hour and saves weeks. With a written brief you can mark each review as relevant, partly relevant, or irrelevant. Without one, every review looks relevant because you have nothing to compare it to.

Include your constraints. If you cannot run user-level data because of your own consent setup, a review praising granular targeting is not describing a benefit you can use. The same applies to minimum spend: a platform praised for accessible entry pricing is irrelevant if your budget sits below its floor.

Write the success measure as a number and a date. Awareness is not a measure. Brand lift above a stated threshold, among a defined audience, by the end of a twelve-week flight, is. Reviews rarely state their success measure, but you should be able to infer whether theirs resembles yours.

Our connected TV advertising selection checklist turns this into a scorable list you can reuse across every vendor you assess.

Step 2: identify who wrote the review and who paid for it

Ask three questions of every review. Who wrote it? What did they get for writing it? What access did they have?

A review by a buyer with no commercial relationship to the vendor carries different weight from a vendor-written comparison that ranks its own product first. Both can be useful. Only one is independent.

Look for a stated method. If the review does not say how products were tested, what period was covered, or how many campaigns were included, the ranking is an opinion. Opinions are cheap, and there are a lot of them.

Access matters as much as independence. A reviewer who spent a week in a self-serve dashboard knows the interface. A reviewer who ran six figures of spend through a managed service knows the account team, the reconciliation process and the invoice quirks. Those are different reviews of the same brand.

A named method is also what makes a review repeatable. Our connected TV advertising review methodology sets out the evidence we expect before a product score means anything.

Step 3: check the sample size and the dates

A review based on one campaign is a case study. That is not a criticism, but it is a limit. One campaign cannot separate a product's effect from the sector, season or creative that ran with it.

Sample composition matters as much as size. Ten campaigns run by the same agency, for the same kind of client, in the same quarter, is closer to one case study repeated than to ten independent tests.

If a review quotes an average result across clients, ask for the spread. An average hides the clients who saw no effect at all, and those are the cases you want to understand before you commit budget.

Check the dates. Was the test run over a seasonal peak, a quiet period, or a period when a major platform changed its inventory rules? If the review does not give a date range, ask.

Where a review gives performance numbers, look for the denominator. A 30 per cent lift on a small base is not the same as a 30 per cent lift on a large one. The brief you wrote in step one tells you which is realistic for you.

Step 4: test audience and reach claims against published data

Audience claims are the easiest part of a review to verify, because much of the data is public. If a review says a platform reaches a certain audience, check that against a named source rather than accepting the figure.

Nielsen publishes weekly top ten streaming and TV ratings that show what audiences are actually watching, which is a useful cross-check on reach claims. The Nielsen Top 10 streaming and TV ratings are free to read and updated regularly.

Be careful about the difference between a platform's registered users and its measured viewers. A review that blurs the two is not measuring what you think it is.

Watch for duplicated reach. A plan that buys the same household through a broadcaster app and through a programmatic route will count that household twice unless the measurement accounts for it. Reviews that quote combined reach without mentioning deduplication are worth less than they appear.

Step 5: examine the creative and format assumptions in the review

Connected TV creative behaves differently from other video. Longer viewing, larger screens, and co-viewing change what works. A review that tests one format and generalises to all formats is overreaching.

Check whether the review tested the format you plan to run. A fifteen-second non-skippable spot is not the same as a six-second bumper or an interactive ad unit.

IAB UK publishes an introduction to creative best practice for digital advertising, including video, which is a sensible baseline when judging whether a review's creative assumptions are reasonable. Read the IAB UK introduction to creative best practice before you accept a review's claim about format performance.

Ad break structure also matters. Thinkbox explains how ad break innovation affects attention and recall, which is worth understanding before you accept a comparison that ignores break design. See the Thinkbox guide to ad break innovation.

Co-viewing is the assumption most often left unstated. If a household of three watches together, impressions and reach diverge, and any cost-per-reach figure quoted in a review needs to say which one it used.

Step 6: check the compliance and regulatory position yourself

The advertiser is responsible for the ad. That does not change because a platform or a review says approval is straightforward.

UK advertising rules are enforced by the Advertising Standards Authority under the CAP Code. The ASA publishes guidance for businesses on what compliance involves, including the evidence you need to hold before making claims. The ASA advice for businesses is the starting point.

Broadcasting and on-demand rules sit in a wider statutory framework. The Communications Act 2003 is the primary legislation, and it is worth knowing which regime applies to the inventory you are buying.

In practice, the regime depends on where the ad appears. Broadcast channels and on-demand services regulated under UK broadcasting rules carry different clearance expectations from ad-funded streaming inventory delivered over the open internet. A review that treats them as one category is not describing your buy.

Clearance times quoted in reviews are usually best-case examples. Build your own buffer into the schedule rather than planning to the fastest turnaround you have read about. Where a review describes approval processes, treat that as a claim to verify. Our connected TV advertising rules and compliance guide covers how the regimes differ across broadcast, on-demand and online video.

Step 7: compare products on your criteria, not the review's

Now do the actual comparison. Take your brief from step one and score each shortlisted product against it. Do this before you read the ranking, so the ranking tests your judgement rather than replacing it.

Use the same scale for every product. If a review gives stars, convert them to your scale or discard them. Mixed scales are how weak products end up looking strong.

Where two products score close, the deciding factor is usually operational: reporting access, contract length, minimum spend, who answers the phone when something breaks. Reviews rarely cover these because they are dull. They are also where most of the pain sits.

Commercial terms deserve a second pass. Ask what happens at the end of a fixed term, whether inventory is bought at a rate card or at auction, and whether fees are charged on media or on total spend. A review comparing headline CPMs tells you nothing if one product charges a platform fee on top.

Our connected TV advertising product comparison shows how to lay this out so the differences are visible rather than buried in prose.

How to read agency reviews without being misled

Agency reviews have an extra problem: the product is a team, and teams change. A review of an agency from two years ago may describe people who have since left.

Ask who actually ran the campaigns. Ask whether the reviewer is still a client. Ask whether the agency had any say over what was published. Ask for two references you can call, and ask each one what went wrong.

Watch for reviews that describe the pitch rather than the delivery. Pitches are performances. Delivery is the monthly report, the reconciliation, and the willingness to say a channel is not working.

Check whether results are reported against the same denominator across agencies. One agency may report return on ad spend against media cost, another against total cost including fees. Two numbers that look comparable often are not.

Our connected TV advertising agency reviews in England sets out how to assess agency evidence in the English market, where most UK agency buying is concentrated.

A simple scoring method you can reuse

Score each product from one to five on audience fit, measurement quality, creative support and compliance clarity. Then score commercial terms and reporting access on the same scale. Maximum score is thirty.

Weight the criteria to match your brief. If measurement is your main risk, double that criterion before you total the scores. If you are buying a small test, weight commercial terms more heavily.

Record why you gave each score in one line. When you revisit the decision in six months, the reasoning matters more than the number.

Finally, set a threshold in advance. For example, a team might decide that anything below twenty-two out of thirty is not shortlisted. Deciding this before the scores arrive stops you rationalising a weak option into the list.

Run the same scoring sheet every time you review the market, usually once or twice a year. Comparing your own scores over time shows whether a product improved or whether you simply got a better account manager.

Common questions

How many reviews should I read before choosing a connected TV product?

Three or four independent sources is usually enough to see where they agree and where they conflict. If they disagree on a factual point such as audience size, go to the primary data rather than reading more reviews.

Are vendor-written comparisons worth reading at all?

Yes, but read them for specification and pricing structure rather than judgement. They often describe features accurately while ranking themselves first, so use the facts and ignore the ranking.

What should I do if a review gives no methodology?

Treat it as opinion and weight it accordingly. You can still use it to find products you had not considered, but do not let it settle a decision that involves significant spend.

Does a high review score mean the product will work for my campaign?

No. A score reflects the reviewer's brief, budget and sector. Your brief may differ on all three, which is why the scoring method above starts with your own criteria rather than someone else's ranking.