Erik Dietrich

#general - January 09, 2025 at 05:43 AM

@Amanda Henry, first of all, I'm sorry for (though not necessarily surprised by) what you're going through. I also don't have any specific wisdom for fighting the good fight, other than exactly what you proposed -- document what's been asked of you, make it clear and visible, and circle back to it each time the powers that be get distracted by a shiny object and change your marching orders. As oblique food for thought, about a hundred years ago in my software engineering days, I used to listen to a podcast called .NET Rocks where one of the hosts was fond of giving the career advice, "change your company... or change your company," meaning, if you can't get a startup with a bottom-up GTM to understand why 3x-ing a reliable distribution channel is a win, maybe you look for a landing spot where you don't need to explain that. Anyway, I hope for your sake it doesn't come to that.

#general - January 08, 2025 at 08:50 PM

IME, it's pretty common for technical founders (I'm guessing) not to understand marketing well enough to manage it competently. Hopefully for your sake, that isn't the case, but "give your report a goal and don't empower them to control its realization" is a smell.

From a purely utilitarian perspective, you could try (semi-) malicious compliance: publish whatever ChatGPT burps out, and invite the founders and engineers to all read the blog post, live on the site. I imagine you'll have an inbox full of suggested edits in record time.

If that generates a kerfuffle, you could ask leadership to force rank your primary KPI and countervailing KPI. "Should I hit my cadence goal, even if it means sacrificing quality, or should I hit my quality goal, even if it means sacrificing cadence?"

I realize that this approach might be a bridge too far and that it would certainly burn some political capital, to say the least. But it might be worth keeping in your back pocket as a hail mary if the situation doesn't improve. A slightly less risky move might be to create only content that requires no input and explain, when questioned, that it's the only lever you have to pull given your current constraints.

#general - December 19, 2024 at 07:28 PM

We do this on an ongoing/realtime basis for probably something like 20-25 sites and growing. This requires a fair bit of automation (ETL tooling and a data warehouse for instance), but here's what I'd personally do in my toolset if a client asked me to do what you're doing.
1. Assuming it's valid, grab the sitemap(s), download the URLs/lastmods into a CSV and paste the results into an Airtable.

2. Have a multi-select field called "Attibutes" that you use to tag the URLs with data points and potentially actions (e.g. the action's in @Eric Doty (Superpath)’s list like "Keep", "Refresh", "Merge", etc.

3. You can also use this for tagging funnel stages, which is probably going to be an easier paradigm given that you might have URLs that blur the lines and tag them with more than 1 stage.
4. Create columns for data that you have in the spreadsheet, merge the results and paste the traffic figures into the Airtable base to help with your analysis.
Here's what this looks like on a demo/community site that I own

#general - December 18, 2024 at 09:11 PM

Not to be flippant, but you could refer them to Betteridge's Law of Headlines and suggest that the best way to grok an article entitled "X is dead" is to conclude "nah," without reading it. Mainframe computers, for instance, have been chugging along steadily in spite of spending almost the last 50 years being dead.
In this case, someone proclaiming X to be dead alongside shutting down their X practice seems like a self-soothing narrative that doesn't demand any introspection about what might have gone wrong with their practice or how they could have evolved it as the market shifts.
(Incidentally, I'm saying this as someone with a fair bit of data into the demand for content, specifically targeting SEO, who has seen demand steadily increase over the last 12 months)

#freelance-talk - November 01, 2024 at 01:31 PM

I would turn it down immediately without trying to negotiate, saying something like "in that price range, you're looking for a different category of service provider than what I offer." If you attempt to negotiate, at best it'll turn into amateur hour at the car dealership where you bark prices at each other and meet in the middle. That will result in you having a client that thinks you're too expensive, and them having a vendor who doesn't really value their business. And that -- you being at the top of their price range and then being at the bottom of yours -- makes for miserable relationships.

#general - October 03, 2024 at 03:17 PM

Personally, the first step I'd take would be to start sending invoices with late fees and notify them that you're going to send them to collections. Also, you don't have to involve an attorney to tell them that they're going to be hearing from your attorney.

#general - October 03, 2024 at 03:16 PM

I'd be careful doing that, or they might involve a lawyer (assuming their reason for non-payment isn't cash flow problems for failing), since what you're proposing sounds like libel, assuming you're going to review-bomb their actual product or service.

#freelance-talk - October 03, 2024 at 03:08 PM

For your specific context, it might be worth asking yourself whether this person is even an ICP and at all knowledgeable about marketing strategies. Those two messages make it sound like the founder's first rodeo.

#freelance-talk - October 03, 2024 at 03:04 PM

This is kind of defacto market research. And when you're doing market research, one person's take is just a singular, potentially very non-representative data point.

This kind of reminds me of when an unhappy customer at a diner goes on Yelp with lots of opinions about how that business should be run, but those opinions ignore any context about who the diner's ideal customer is and how their business model works and are just hyper-optimized for how to please that person: "if they don't provide the half gallon of free ketchup I drink with every meal, they're going to go out of business!"

#freelance-talk - October 03, 2024 at 12:25 AM

Huh. I do think having some kind of intake form or specific set of questions during discovery could help, but it's kind of wild that people very used to scoping concerns around content would do that. In situations like that, it feels a little more like a not-the-best-faith attempt to negotiate after the fact.

#freelance-talk - October 02, 2024 at 09:25 PM

Anyway, a little rough around the edges, but the idea is to create a process that allows for them to articulate to you what's important to them, even if they don't immediately know how to do it and it takes the form of "I can't define it initially, but once you submit something I'll be able to express it from there."

#freelance-talk - October 02, 2024 at 09:24 PM

I'm looking at this from the client's perspective, and it seems like you might be scoping with variables that are more important to you than them. IOW, it sounds like the word count and number of links are important to you and topical coverage and (I guess) comprehensive linking are important to these clients. This is a bit of an extreme analogy, but it'd be like going to get your car washed and them telling you that it's 5 cents per square foot of car washed with up to 5 spins of the buffer drying thingy. The car comes out and the tires are still dirty, so you ask if they can clean the tires. They say that it's out of scope and they'll need to charge you another 60 cents. You'd probably say, "oh, sorry, I guess, that's fine, I just want my whole car clean."
I might be off base if your clients are seasoned in requisitioning content and very familiar with the way freelancers tend to scope it. But it could be that they're not really even particularly aware of the word count or linking specifics. If that's the case, or you suspect it might be the case with a prospect, you could consider offering some kind of prototype run in the first piece of content or two where you specifically bake in an additional charge as a kind of paid discovery. "It's common to charge per word, flat-rate in bucketed ranges, but 9 times out of 10 it takes a few deliverables before they fully understand and can brief out what they want to see in a 1,500 word piece. So the initial deliverable will be 2 pieces of content and a briefing process we can operationalize to ensure you get what you need within the agreed-upon scope."

#freelance-talk - October 02, 2024 at 04:27 PM

If the clients push scope in a way that's a pattern, you could present the expanded scope to them as a first-class option when making proposals. In other words, if standard scope is a single round of revisions and clients often ask for a 2nd round, you could present both options to them during sales calls and specify that they can upgrade ala carte per article. Then you've set the default to be an up-charge and can respond without needing to stress over messaging and context -- just an automatic, "sure thing, I'll revise the invoice and get that over to you."

#general - October 02, 2024 at 04:16 PM

So, tl;dr, I'd tell your stakeholder that the project can only move forward if the people they're soliciting feedback from provide actionable feedback, rather than playing Siskel & Ebert

#general - October 02, 2024 at 04:16 PM

And that's an awful process.

#general - October 02, 2024 at 04:16 PM

"This is jargon" can only be operationalized by continuously submitting things to that person and hoping that after a while they stop deeming what you submit "jargon."

#general - October 02, 2024 at 04:15 PM

In other words, feedback can be specifically objective and actionable or it can be subjective. If it's objective/actionable, it can be operationalized and you can make progress. For instance, "we consider words X, Y, Z to be jargon and don't want them in the copy." Easy enough, both to implement and to work into a standing style guide.

#general - October 02, 2024 at 04:14 PM

The solution, I'd say, is to categorize the feedback.

#general - October 02, 2024 at 04:14 PM

Non-invested stakeholder review always seems like a benign throw-in, but it's really a subtly bad incentive. Your stakeholder is, presumably, the one on the hook for the homepage rewrite. Product marketing, well, isn't. So your stakeholder de-risks the situation by asking product marketing for a review, and people asked to review something will almost invariably weigh in with critical feedback, lest they seem uninformed. The homepage not being updated isn't product marketing's problem, so providing vague, non-actionable critical feedback is no problem at all for them. The way to combat this, IME, is to make product marketing's feedback your contact's problem (i.e. it's what's holding up the goal).

#general - September 10, 2024 at 04:47 PM

I might be careful pursuing this any further, especially if the ongoing relationship with the client matters as much as her testimonial hypothetically matters to your sales process. You asked for a favor and she politely said no. If you're misreading the situation even a little (e.g. she's just being polite about the reason, and citing what seems to be a no-fault, incontrovertible blocker even though she has other reasons) you run the risk of actively annoying her. In other words, if she is truly eager to give you a testimonial, but there's some stupid rule that prevents her from doing so, then she might be receptive to your solutioning at how to get around the rule. But anything less than that and now you're essentially arguing with her.

#freelance-talk - September 04, 2024 at 09:49 PM

Revenue is kind of a context-free metric. If I were a freelancer, I'd set a revenue goal only as an enabler for some kind of more concrete goal. For instance, if I wanted to evolve my practice into an agency, I'd set revenue goals based on expanding profit margins to the point of sustainably hiring. If, on the other hand, earning more revenue would enable more things on the lifestyle or personal finance side, I'd let whatever I was trying to do (new house, bucket list vacation, etc) determine incremental revenue goals, especially if I was just taking distributions and paying self-employment tax, rather than a salary. If last year went well, but I wanted to also take a $5K vacation, then I'd shoot for an incremental $8K in revenue (accounting for tax). Repeat until I'd added in all of my Wishlist items but wasn't signing myself up for a brutal schedule.

#freelance-talk - August 23, 2024 at 11:53 PM

(Kinda my own fault -- I still accept connects from strangers when it isn't obvious to me what the person is going to pitch, and I should probably stop)

#freelance-talk - August 23, 2024 at 11:52 PM

Just to illustrate my earlier point from the recipient's perspective, I'm going through my periodic LI message inbox triage right now, and this was the very first unread message thread. Is this person:
A) So overwhelmed with free time that they're looking to make friends with everyone that ever attended my alma mater or...
B) {sigh} acting casual while trying to chum the waters?

#general - August 23, 2024 at 07:42 PM

Sometimes just making it visible is enough. If they see that they're adding 2 weeks on average to your publishing process, they'll be naturally gamified to reduce that figure.

#general - August 23, 2024 at 07:41 PM

Drawing on something from my consulting days, I'd suggest making the impact "big and visible." In other words, if your team has an editorial calendar you could create a view of that with some kind of metric or metrics like "total legal review days" or "legal average lead time" or, perhaps best of all "approximate views (or whatever) lost to legal review". Publish that metric, report on it regularly, distribute it to your management...
Legal is a cost center, and you're a profit center. Someone with pull in both worlds will eventually notice and start to care.

#freelance-talk - August 23, 2024 at 01:19 PM

YMMV, but I'd be careful with replacing "connect 'n' pitch" on LI with "connect, lets-be-friends, pitch." As someone on the receiving end of massive amounts of both tactics, I know exactly what's coming when a stranger connects and says "I just want to get to know you, what are some challenges you're facing?" Instead of a pitch, it's going to be a bunch of pen pal stuff I don't have time for... then eventually a pitch.

I'm not saying this under the delusion that I'm necessarily representative of everyone you might pitch to on the medium, nor am I saying it can't/wouldn't work. What I am saying is that the "let's be friends so I can later pitch to you" is ubiquitous on the medium, so you'll probably need to be very careful that it doesn't come off as inauthentic.

#general - August 22, 2024 at 03:09 PM

Good news is, if you're talking about an oops window of 2 months, you'll likely be able to recover nearly all of your traffic and downstream KPIs fairly quickly if you sort out your URLs fairly soon. Index status is surprisingly durable in the short term

#general - August 22, 2024 at 03:08 PM

FWIW, you're far from alone. People come to us a lot with "what happened to my traffic" and the most common two things are that they somehow unpublished or deleted content or they unwittingly did something with adverse consequences during a site migration.

#general - August 20, 2024 at 10:25 PM

Once you've isolated the specifics of the loss and formed some hypotheses about remediation, you can systematically test them out.

#general - August 20, 2024 at 10:25 PM

That can help you start to triage. If the loss is uniform across a lot of traffic sources, it's more likely to be something systemic. If the loss all comes from 1 or 2 URLs, there's probably something wrong with those URLs.

#general - August 20, 2024 at 10:24 PM

It shows which URLs have accounted for traffic loss over the last (in this case) 30 days.

#general - August 20, 2024 at 10:24 PM

The first thing I'd do is divide and conquer. Your content categories and URLs are probably not all losing traffic equally. For instance, this is a dashboard view that I maintain (example on a toy website I own):

#freelance-talk - August 20, 2024 at 10:14 PM

(I don't think the latter is likely, but both of you brainstorming and laying it out would help you both think it through)

#freelance-talk - August 20, 2024 at 10:14 PM

Depending on the end-client and how they do things, it could be that 15% is way too much for them to take, but it also might actually be too little if, for instance, all of the stuff they do on your behalf costs thousands a month in insurance and legal

#freelance-talk - August 20, 2024 at 10:12 PM

What I'd suggest doing, if you have a good relationship with the agency, is having them list what, exactly, they're doing on your behalf because it might be things you're not thinking of. I say this as someone who has navigated enterprise legal and compliance then turned around and subbed out work, and done so in situations where it would be literally impossible for a freelancer to work with the enterprise. Here are some things they might be doing that you're not thinking of:
• Paying you Net-30 and receiving money Net-90 (so they're literally lending the client money for your services and absorbing non-payment risk).
• Carrying insurances that you don't have.
• Absorbing legal exposure on your behalf.
• Handling ongoing compliance requirements that potentially change frequently (meaning, they spend weeks and $$ on legal every time the client decides to add some clauses to its MSAs so that you don't have to).
• Bridging location/payment requirements that you wouldn't meet.

#content-collab - August 16, 2024 at 07:45 PM

Sure

#content-collab - August 16, 2024 at 06:11 PM

You're asking if I use URL rankings for keywords to support "backlink analysis?" I'm not sure what you mean about backlink analysis. Like, have we ever run a study to see if we can corelate backlinks to higher rankings?

#general - August 16, 2024 at 02:42 PM

We use Data for SEO for a bunch of things, one of them being looking at rankings for a URL against its primary keyword. The main goals with this is to monitor performance for the URL against the performance we'd expect for it (is it ranking better, worse, or about where we'd expect) and to identify leading indicators of traffic loss to schedule refreshes.

#general - August 15, 2024 at 06:46 PM

So I'd make his site the model for where to hunt for relevant audience on the topic. The site/publication would talk about a much broader topic, but have a bit of relevant content to those concepts.

#general - August 15, 2024 at 06:46 PM

If your client provides a solution in or around those technologies, I'd recommend focusing on the most popular languages/stacks of your existing users. Those tools are pretty tactical in the grand scheme of things, and probably only interesting to most engineers at the time of deciding whether/how to use them, rather than on an ongoing basis. To give you an idea of what I mean, consider Eugen's site, Baeldung (he probably still accepts sponsorships, BTW). The site cranks out a ton of content in the Java world in general via guest blogging, but if you look at the content related to grpc, it peters out around page 2, and I imagine that's one of the heavier treatments of the topic.

#general - August 15, 2024 at 03:25 PM

One very popular newsletter that also has other sponsorship options and a mature media operation is Last Week in AWS.

#general - August 15, 2024 at 03:23 PM

My take on this would be kind of skewed based on people that I know and that used to feature my content. Is there any particular tech stack you're looking for? Almost all of the publications out there, with the exception of stuff like soft skills or learning to program, will be segmented into stack or discipline (e.g. a newsletter for Python developers or a newsletter for SREs)

#general - August 15, 2024 at 12:11 AM

It's a pretty broad topic to take a whack at in a thread, but a few thoughts.
1. Self-promoting/distributing in venues like Hacker News, Reddit, Stack Overflow won't go well. Ditto private developer communities in Discord. The name of the game there is to get others to promote your stuff organically, and that's kind of a long game, especially for a brand.
2. Especially if you're not targeting SEO with the content, you can distribute to DEV, DZone, Hashnode for free, as well as some other more niche communities like Web Developer.
3. If you pay you can have deals with those community sites that get you a lot more attention (e.g. DEV and their community challenges). You can also look at venues like the New Stack and Hackernoon where you can distribute content to that audience for pay.
4. If the content is mainstream developer audience (as opposed to overly niche or targeting a specialized IC like an architect), SEO is a channel where you can simply, though not cheaply or always easily, build 6+ figures per month of largely qualified traffic.
5. If you want to pay, I'd think a little outside of the normal box since that audience is extremely good at igonring ads and blocking them with tools. Developer community influencers usually don't know very well how to monetize audiences when they have 100K or fewer subscribers, so you can do some pretty good value buys there.
6. Ditto (value) in a lot of niche developer podcasts looking for sponsors. For instance, I've done a number of placements with Chuck, who runs this podcast network and it's a good deal on CPM for that audience. I've got a database of some others as well.
7. Sponsoring newsletters is probably a good tactic, though I'd recommend discussing (and probably deferring to) the newsletter author on how to position you as a sponsor.
8. This is a less common tactic, but a lot of developers will start blogs or YouTube channels or whatever and kind of abandon them or let them peter out. If a channel like that is attracting visitors organically, you could probably buy them out inexpensively and it would be found money for them.
9. YMMV, but I wouldn't target LI as social media for developers. IME they mostly view it as a silly place. Twitter is (or used to be) where you'll find more natural channel participation.

#general - August 14, 2024 at 01:21 AM

One thing I'm curious about is that Templr tool that you listed, @Ryan Levander. Does that act sort of similarly to something like VBA in office applications, and enable client-side automations with Javascript? Their video seemed to be mostly about formatting components with CSS/HTML.

#general - August 14, 2024 at 01:14 AM

FWIW, my objections to it aren't related to what you can build with it but are more about configuration management. But if it works for you or anyone else, by all means, enjoy and be productive.

#general - August 14, 2024 at 01:11 AM

¯_(ツ)_/¯

#general - August 14, 2024 at 01:11 AM

Sounds like your mileage, indeed varies.

#general - August 13, 2024 at 09:46 PM

I've been absolutely immersed in this series of techs lately, integration GA, GSC, Airtable and some other tools in a BigQuery data warehouse and then using Looker (for the moment) to report on it. This is with the intention of turning it into an offering. Anyway, here are some recommendations in no particular order:
• I wouldn't worry about learning "SQL" and instead just focus on how to query BigQuery, itself. (It's an imperfect metaphor, but learning SQL is kind of like learning Latin before you learn a romance language)
• The actual schema behind GA4 is pretty rudimentary, so I wouldn't spend a ton of time on even learning BigQuery itself beyond relatively simple stuff that ChatGPT can easily tell you how to do.
• If I were you, I'd set aside Looker at the beginning, and focus on how to ingest the data BigQuery and then setup (and possibly version control) views of the data as you develop those. You'll be able to ingest them into Looker Studio later pretty easily when you're ready to report, but I'd focus on learning to stage the data you want first and then turn your attention to the user interface when the data is good.
• YMMV, but Looker Studio is kinda trash IME. You might want to look around at other options (I'm going to kick the tires on Metabase when I'm ready to start charging for what we're doing)
• Just to get an early end to end win in the books, I'd recommend making your initial goal just to recreate standard GA4 reporting via BigQuery and SQL views. That way you can validate that you're doing that stuff properly by comparing against a known commodity.

#general - July 16, 2024 at 07:24 PM

If you own the agency, you could check out the Build a Better Agency community. It's a Facebook group, though Drew, the founder, also has a podcast

#general - June 25, 2024 at 01:14 PM

And that's genuinely my motivation. With content, for instance, there is a subset of people that will literally never be happy outsourcing it, and if that's the root cause of their dissatisfaction with a previous vendor, much better to know that and steer clear.