Agency is pitching a high-budget UX audit: What is it exactly and is it necessary?

We paid 280,000 RUB for a similar audit on our shoe store last spring. The report revealed that our one-click checkout module, which is huge across Russia, was getting hidden behind the keyboard on mobile. Just fixing that area and the address autofill box noticeably slashed our cart abandonment rate and made our money back in two months.

Many agencies on the market sell so-called UX audits that are just automated software exports mixed with generic design best practices copied off the internet, all packaged into a 40-page PDF. To see if they actually do deep-dive work, ask for an anonymized sample audit report they've done before. If it’s just packed with empty theory, definitely stay away.

Before the audit even starts, they need to ensure your site's analytics tags and event tracking setup are fully in place. If form drop-offs, scroll depths, and error triggers aren't logged properly, the agency won't see any real data; they'll just be making visual guesses. If they aren't assessing data health during the first week, that audit is incomplete.

Ask the agency directly: will this audit include one-on-one usability testing with actual customers on our site? If 350,000 rubles just gets you their in-house designers sitting around reviewing the site, it’s not worth it; they need to assign real purchasing tasks to at least 5-6 people from your target audience and record their screens.

cart abandonment sometimes isn't even a ux issue, it's just surprise shipping fees popping up at the last step or vague delivery dates. like imo before throwing money at an agency you should check your checkout funnel data in analytics yourself first.

The success of an audit doesn't lie in how pretty the report looks, but in whether you actually have the developers to implement it once it's delivered. Even if the agency delivers an amazing UX audit, if you don't have the technical resources to execute the suggested UI changes, that 350,000 RUB will just rot on the shelf. Discuss the post-report implementation effort with the agency upfront.

Did they include testing consultancy in the contract? Do they suggest pushing the new UI variants from the audit straight to production, or will they run them in parallel A/B tests against the old design to validate the data? That distinction will tell you everything about the quality and sincerity of their proposal.

absolutely... anyway if I were to add anything: Dont hesitate to ask; tohse who dont ask always pay more.

if you post the result here it will help others too.

You're right, I've been down that road too. Hasty decisions become decisions you have to fix six months later.

I'm also curious if anyone does it differently.

I've been down this road, let me tell you. Ask who prepared the quote and who will actually do the work.

If it's your first time, start small; scaling comes later. Of course, it varies if your situation is different.

Exactly like that. The system lives on after delivery; maintenance is a separate line item.

In legacy systems, the most expensive thing is unknown dependencies. Of course, it varies if your situation is different.

I agree.

I went through the same thing.

I disagree with you on this point. If acceptance criteria aren't written, when the work is done is open to debate.

Code without setup documentation isn't yours, even if you have it. Proven by experience.

We've heard this a lot but it never happened like that for us. Don't hesitate to ask; those who don't ask always pay more.

Implementing a change request process doesn't slow things down, it speeds them up. Of course it varies if your situation is different.

There is something to watch out for. like hasty decisions become decisions you have to fix six months later.

If I were you I'd go this route.

Correct. Start with a small trial; don't commit to everything at once.

Everything goes well for the first three months; problems arise in the fourth. Just leaving this note, it might be useful.

correct.