CoreDash vs DebugBear

I built CoreDash, so I'm biased. Keep that in mind while you read this. DebugBear is one of the better performance tools around, and in a few areas it's better than CoreDash. I'll point those out.

If you already run DebugBear RUM and it answers your questions, you don't need to switch. You're measuring real visitors, which is the important bit. This page is mostly for people who are still choosing, or who are running into the limits of what they have now.

What DebugBear does well

The big one: lab data, field data and CrUX in one place. Scheduled Lighthouse tests run from more than 30 locations, with network throttling at the system level (not simulated), several runs per test and the median kept. You can test user flows and logged-in pages, and you get before/after videos. When you sign up you also get about 40 weeks of CrUX history for your site.

debugbear.com, October 2026.

The RUM attribution goes deep. LCP is split into the four subparts (TTFB, resource load delay, resource load time, render delay), INP into input delay, processing time and presentation delay, and TTFB into redirect, wait, DNS, TCP, SSL, response and download. For INP they use the Long Animation Frames API to show the script URL and invoker type, event listeners included. The INP target tables even have screenshots of the elements.

They have conversion tracking as well, which ties Core Web Vitals to bounce rate, session duration and checkout starts. If you need to convince someone to pay for performance work, that's the chart you'd show them.

Other things worth knowing: the snippet sets no cookies unless you turn on session tracking, and it doesn't store IP addresses. They're SOC 2 certified, have a REST API, a CLI and webhooks, and an MCP server since June 2026. Their blog, docs and free tools (the INP Debugger for example) are some of the best free web performance material out there.

RUM vs Lighthouse and CrUX

Lighthouse and PageSpeed Insights load your page once on a simulated phone. CrUX only counts opted-in Chrome users, and a page only shows up once enough of them have visited it (see not enough data in Search Console). RUM measures every visitor in every browser. DebugBear and CoreDash both do that well.

Most sites I audit don't have any RUM. They work through the Lighthouse list (Minify CSS saves a few kilobytes, no visitor will ever notice) and then wait for the 28-day CrUX window to roll over to see if it helped. So if it's DebugBear or nothing, go with DebugBear.

CoreDash and DebugBear side by side

Pricing and features checked September 2026. Prices are per month, billed annually, in USD.

Where I'd pick CoreDash

You only pay for field data

DebugBear sells RUM together with lab testing. If you use the lab tests, that's a fair deal. If you only want field data, the $249 for 500K page views also pays for 10,000 Lighthouse runs you won't use. CoreDash is $31 for the same 500K pageviews.

DebugBear pricing page, October 2026. RUM starts on the $99 Startup plan.

And the difference gets bigger as you grow. Your RUM volume goes up with your traffic, the number of lab tests you need doesn't. On DebugBear, once you go over your quota RUM gets sampled and then stops storing new data, unless you buy add-ons (+500k page views for $60 a month). On CoreDash every plan has every dashboard and unlimited tracked pages. Moving up a plan buys you volume, sites and retention, the features stay the same.

CWV Superpowers writes the code fix

Both tools have an MCP server now (DebugBear's arrived in June 2026), so Claude, Cursor or whatever agent you use can query your RUM data in either. That alone doesn't decide anything.

What CWV Superpowers adds is what happens after the query. It's a free, MIT-licensed agent on top of the CoreDash MCP server. It pulls your field data, finds the bottleneck that takes the biggest share of the time, and then records a Chrome trace with mobile emulation, Fast 3G and a 4x CPU slowdown to find out what causes it. You get a code fix with file, line and diff, or an HTML report if a developer is going to do the work. DebugBear's fix prompts start from their test results, which is fine if you run their lab tests. I start from field data, because that's what Google ranks on. A lab run shows you what was slow on one simulated phone. Field data shows you which templates are slow at p75 on mobile, for the people who actually visit.

The MCP server is included in every CoreDash plan, Starter too. It's the same endpoint as the API and it's read-only. There's no write path, so handing an agent the key can't change anything in your project. AI Auto Alerts use the same field data: they check hundreds of segments every day and only flag a regression when it hits at least 2% of your traffic. Fixing INP with an AI agent walks through it on a real site.

Agencies: more sites per plan

DebugBear's Startup plan has 1 project and Team has 5. Need more client sites and you're looking at Corporate at $699 a month, or +10 projects for $50 a month. CoreDash's agency Studio plan is $71 a month for 10 client sites, 1M pooled pageviews and unlimited users. Growth is $159 for 25 sites and 3M pageviews.

For an agency I'd look at the Agency API before the price. One master key creates and removes client projects over REST, and the same three MCP tools can read any client project by ID, so an agent can go through all your clients in one run. CoreDash isn't white-label though. If you send clients branded PDFs, DebugBear's white-label exports (Corporate and up) do that. With CoreDash you get the data and you build your own reporting on top.

Retention

On the entry plans DebugBear keeps more history: 18 months on Startup, against 6 months on CoreDash Starter and 1 year on Standard. On the bigger plans it's the other way around. DebugBear's self-serve plans max out at 24 months. CoreDash Studio keeps 3 years, the agency plans 3 to 5 years and Enterprise 8 years. If you want this Black Friday next to the previous two, 24 months is too short.

I didn't put attribution in this list. DebugBear's is very good and there's no big gap. There are a few CoreDash dimensions I'd miss in other tools though: LCP priority (preloaded, fetchpriority high, or lazy loaded), INP load state and device capability. The Diagnosis view checks for 14 LCP anti-patterns and gives fixes for your stack. LCP priority is the one I use most. One column tells you someone put loading="lazy" on the LCP image, and removing it is a one-line change.

When to pick DebugBear instead

• You want scheduled lab tests, RUM and CrUX history in one tool, with before/after videos of the lab runs.
• Procurement needs SOC 2.
• You send clients white-label PDF reports, or your team works in Microsoft Teams.
• You already use DebugBear and it shows you what to fix. Then I wouldn't switch.

Try both

Both have a 14-day trial, no credit card needed. Put both on the same site for two weeks and see which one you end up opening. Want to see the dashboards first? There's a CoreDash demo, and pricing shows what your traffic would cost. Comparisons with other tools are on the overview page.

Sources: DebugBear pricing, DebugBear RUM, RUM setup, RUM data privacy, additional RUM metrics, DebugBear MCP server, DebugBear for agencies.