Ask HN: Fastest way to get recurring revenue with hosting?
I like startups and creating webapplications. But my main income (after working hours) is still selling websites / hosting or invoicing packages (=> customization of the invoice layout, configuring webapps, ...), which is like 85% of my income). The rest is mostly from hosting (webapps / websites)
I'm wondering, what do you sell to people (webapps, websites, cms), which contains hosting and is fairly easy to create a recurring revenue with?
And do you upsell? (eg. selling a website + email marketing application + affliate gmail for biz)
39 comments
[ 0.26 ms ] story [ 971 ms ] threadWhen it is a new company (is going to launch in 1 month for example), i propose them to set a launch page (basic one, created in max. 15 minutes) for 100 €, to collect emailaddresses.
The launch page includes a text email to all of your visitors when they subscribe and say that this proposal doesn't include HTML (for images), because that is custom work and more difficult.
When the moment arrives, i ask them for the text they want me to send to their visitors.
In 70% of the cases, they ask to include a picture of the team.. I explain them that this was not part of the deal, but that i can change the message to a HTML email for 80€ (if provided the assets first).
So, selling a website earned me another 180 €, a happy customer (the launch page is added publicity)
How do you upsell?
I suppose you don't do services to clients? The most important part is that you don't undervalue your own services. See here some opinions of patio11:
https://training.kalzumeus.com/newsletters/archive/saas_pric...
https://twitter.com/patio11/status/463934642500804608
Do not price software at COGS! :-) - Perhaps the most important lesson in any development / selling gig. (i had to learn it the hard way, initial discussed price was too low, had to put a lot of more hours in to get even below minimal wage - once but never again)
It is not immoral in any way. It's such a common practice that customers already expect it.
If they want a HTML mail, they have to pay. They know it in advance. What the customer realizes on the end (like they do in every project), is that they want a picture of the team/building included.
Changed requirements of the project => different price...
In total I could imagine that 100 euros covers up to about 10 hours work if the client wants to change a few things along the way. I don't imagine your time is only worth 10 euros an hour.
This is a fundamental problem with scaling a consulting business - it might feel like you've earned a lot of money for a tiny amount of work, but the reality is that all the time you spent getting and managing that work means it's not very valuable.
It takes about 10 minutes to persuade (in experience), i choose the layout and ask for a picture and a small title + subtext. There isn't a lot to change on the launch page (the layout is almost solely the picture in full)
The changes by the client are usually for the website, but that's a different package. There i create max. 3 first page designs, then i create the whole website. (included according our talk). Then for reasonable changes, i check wether it's part of the package or is it stuff that we didn't talk about that requires >20 minutes of work (eg. new features)... I'm quite clear on that upfront.
The launch page is just an extra income that takes max 40 minutes (talk + create + publish + text email to subscribers).
Then again, i'm also waiting to raise prices untill i get (more) busy with projects to create.
I really hope some people share their knowledge/experience. The more people that share, the more people you can use it to gain a better profit :)
Call it cloud, thats where the money is.
Here's what worked for us:
- after selling a website, we sold the mandatory support and maintenance contract. We considered this service the foundation, or level one of our recurring revenue stream. This is fairly easy to sell and renew. The recurring revenue we got from this was enough to keep us afloat.
- after selling support and maintenance, we upsold the client to "levele two": services which grow our client's online business. The types of services we offered in this plan: everything that needs to be done to reach client's business goals, and that we could deliver well. This was harder to sell (because the type of client needs to be just right for this kind of service), but the amount of money coming in every month is substantial. This is what makes the agency grow in long term.
Here's what didn't work for us:
- web hosting. We've been offering this for more than a decade and in the end, all things considered, it is just not worth it. A combination of support + maintenance + growth-oriented services is a much better bang for the buck. We sold most of our servers and hosting accounts to a specialized hosting company and focused on what we did best.
For the exact details about building, pricing and selling support and maintenance services, check out my book: https://www.simpfinity.com/books/recurring-revenue-web-agenc... (the part about growth-oriented services is coming soon, matter of weeks)
I love talking about the subject of recurring revenue, it's a passion of mine. I'll gladly answer any questions you might have.
Edited: typo.
Currently, a lot of my clients don't need the monthly support option. That doesn't seem to sell very well, because they mostly have no problems.
What can be sold is the support for 3 months (during the start)... But that is where the most problems are, so i'm wondering if it's financially interesting to do that :) . I also mention that i'm a busy guy, but with a monthly support fee. They can call dibs on me anytime.
In my experience, it made no sense to argue with the client about maintenance when the website was "sick". All the client wanted was for us to fix it. I knew that, so I made sure I sold them a maintenance package in advance, when they "didn't need" it. A website is just like a bridge: if you build it, you're responsible for it, but it's not for free. Everything I put in the world, I am responsible for, but the client needs to pay for it.
Here's how we sold support: - 30 days guarantee period (after launch) - we were selling only annual contracts. A lot happens every month. Frameworks and libraries need updating, 404 links happen, browsers update, CMS systems get new security updates... There's a ton of things to keep an eye on as a developer, and you could create an annual plan which your client pays for monthly (give her a discount when she signs a contract).
So you actually asked about selling services and apps hosted in the cloud? That is a good business, if you make it your core business. For example, we resold Google Apps to clients. Since clients were local and preferred to call our phones for support, we sold them our support services too. So, the client bought Google Apps for Business licenses through us and bought added-value services like support, installation, migration, consulting etc.
That's a good business. Specialization and focus helps here, so that you don't spread yourself too thin.
The client didn't care, they just paid us for not having to worry about it and not having the skills the buy the correct config, size etc.
Surely every web dev should do this, unless the client already has a perfectly fine hosting package?
Just never host client emails.
In my experience, small agencies (up to 20 people) usually cannot afford or justify giving substantial resources to web hosting. My agency is now fairly small (five people), and web hosting is so complex and fast-growing, that we decided not to deal with it. We found a good partner a few years ago and we refer some of our clients to them. Some of our clients do pay hosting to us, but it's more of an application hosting type of service (they rent apps from us and pay annually). We have our own dedicated server on which we only allow websites we've developed ourselves. This is a legacy thing: we've been doing that before and we're still billing some of our past clients for app hosting, but stopped selling any new web hosting.
If I was starting an agency from scratch today, I would steer away from web hosting and let the specialized partner handle that for my clients. There are simply too many things that could go wrong if a small agency does not have the necessary expertise. One wrong step and you can lose a six-figure client over a $10 domain. And since there are too many things to keep up with in your core business (web development), your time is better spent learning about your core business, not about the web hosting business.
I definitely am not trying to argue any of your points, but rather chime in with some of my experience.
Except one thing.
DreamHost staff are generalists. They're probably good at fixing hacked WordPress blogs, but they will never be able to compete with you when it comes to in-depth troubleshooting of the exact application that you built for your client. At best, all they can do is direct your client back to you. At worst, they'll misdiagnose the problem and damage your client's website.
You, on the other hand, are a specialist. You know the website inside and out. You can take one look at an error message and figure out exactly which line of which file is causing the issue. You know when the software stack will need to be upgraded, and you know which parts are the most likely to cause trouble after an upgrade. You know when the client is expecting traffic spikes, and you know that when that happens, DreamHost is likely to suspend your client's website.
Your hosting package, should you choose to offer one, must take advantage of these differences. It should be part of a long-term support contract, not a standalone product, and it should be massively overpriced, like, at least an order of magnitude more expensive than the off-the-shelf equivalent. In exchange, the client gets a server stack that is perfectly tailored to their app (nginx, node.js, redis, you name it), a guarantee that they will never receive a canned answer in response to an urgent support request, and a guarantee that their website will not be suspended in the middle of the biggest marketing campaign of the year.
And of course you should be ready to fulfill such expectations. Don't use cheap servers to host your clients, get some Linodes or Droplets instead. There will be no in-house email hosting, it should be outsourced to Google Apps or some other company that specializes in email. Don't mess with cPanel, your clients can call you if they need to make any changes. Everything should be premium-grade, because there's no money to be made in the low-end market. Make your customer feel like your offer is actually worth the combined cost of hosting and support that you're charging them for.
When you offer "hosting"... do you offer intelligent systems? advanced low level networking? CDN? anycast DNS? disaster recovery plan? peak resistance? awesome monitoring system with 24/7/365 support of the solution? 0day level security solutions? nanosecond performance? 99.999% SLA? penetration testing as proactive maintenance? multidevice testing of every change or patch? development, staging, validation and production environments? storage engineers? database tuning? project road-map with weekly (or daily) reports and meetings of a team of engineers analyzing infrastructure usage, logs, new threats, proposals and evolutions? an awesome web interface for ALL customer facing controls? a problem free experience?
Or are we talking about cheapo domain+cert+shared resources "online presence"? If yes, than maybe just stick to one provider and seek for a "reseller plan", to minimize costs, and as said in other comments, start offering a "maintenance package" as part of the products/solutions to get some recurring revenue.
When you get a great team of operations and support engineers with outstanding knowledge and passion for "systems", they stay motivated, and they are not mismanaged, you will be able to monetize them (and their "toys") with "hosting", "cloud", "online presence", "services" or whatever name, in team with the rest of the solutions you sell.
Infrastructure is a complex and expensive topic. If you do it properly, you can move money. You just need more customers wanting your system solutions/team, than the cost of it.
Otherwise, there is many competence and "third party" services, and the average position is to re-sell that, and focus on the ego of "i'm a designer/coder, systems is a second class stuff I cannot convince you to payme more for that".
We did hosting (dedicated and virtualized servers), managed services, and re-housing, this is, rent again the racks/power/IPs which we did rent on different providers.
If I'm true, I think it did work very well because the company did get very big projects, because the company management did have great agreements and contacts with high environments (local government, local banks, football teams, newspapers, etc).
That is, I as a technical person, do not see myself starting from zero the same AND getting the local customers we did use to get. And much less global customers.
I can assure the systems part of the project budgets, did make great money. Even more profitable on maintenance status. The worse on the company was sometimes, the development team's estimation and execution balance.
I'm a fan of bundling — which sounds a bit like what luckyisgood was getting at.
Every consultant wants diversified and/or recurring revenue. This is why just about all of us inevitably create (or try to create) products of our own. Eventually, many consultants get wind of the idea of retainers, which can have the predictability of SaaS but without needing to build and market software first.
The issue arises with how most consultants put together retainers. It's usually something like "I'll sell you in advance 20 hours a month of my time for $2000."
Here's the problem:
Any first grader can figure out that you're effective hourly rate is $100, which is probably less than your real rate — but hey, it's a retainer and it'll relieve your need to always be selling, so that's OK for most.
Since you'll be making $100 an hour on this retainer, your income potential becomes constrained (you're now on the hook for 20 hours a month @ $100/hr) and the client knows what your hourly rate is. "Brennan, I need more this month. I'll pay you $2500 for 25 hours" or "Can I just pay you $100 an hour when I need you?"
And this is where the retainers of a lot of the consultants I've talked with go south, and the relationships sour.
A better approach (which is something patio11 and I talked about during an event we hosted last year) is to instead sell bundles — which could include your time, and hosting — and make these bundles really tricky to divide.
I could sell a client on:
- Hosting
- Backup management
- Framework / security updates
- A/B test experiments and management
- Up to 20 hours of upgrades and modifications
Now it's not so easy to divide the invoices I'm sending my clients monthly by X.
And I could charge... $5000 a month for that. Or whatever would make it so that my client gets both the peace of mind they're looking for (smart guy managing hosting, backups, security issues, etc), a product that's becoming more valuable (running a/b tests, analyzing their funnels, etc), AND a pool of time for me to do whatever random updates they need.
- Security updates + framework
- A/B testing
- 20 hours of upgrades and modifications upfront (charging per hour works or naming it differently and including it in your price)
- Monthly support
What i can sell more easily:
- Backups (else, they have to do it themselves)
- Hosting
- Webapps (email marketing, invoicing, ...)
- Google Apps for business
(Re)selling hosting, webapps, etc. situates you as a middleman; The margins are much better if you not not only sell them on an Optimizely account, but run it for them also.
Sure, people could go out and spend the time to learn just about anything. In practice, they won't, because that's competing with all the other anythings they could be doing.
Sell the thing that actually has significant barrier to entry.
You take care of the easy thing too, because making the client spend meaningful attention on something that would take you five minutes is just silly, and you're there so they don't have to worry about that stuff. But sell the hard thing.
I don't specialize in anything particular. I do both web and desktop apps for my client. They are a small suite of systems for collecting certain types of physics data, mapping it, and performing a basic analysis of subsets of that data. Other than setting up the servers (system administration is a weak spot for me), I've built everything of consequence in the project: from designing the database schema, to implementing and even improving the client's proprietary algorithms, to building a smooth, intuitive (as intuitive as this can get) UX around Google Maps. But it's mostly done now and I'm bored with the project.
Any tips on how to get out of such a rut?
Let's say that your one current client is like having a day job. It's wise to keep them if they're your only source of income, until you find more exciting and equally / more profitable sources of income.
A couple of ideas:
- try to get a couple of new clients, but only a few, so that you can manage them. Put them on a retainer and service them yourself for starters. Repeat until you have enough money from that on retainer. Aim for fewer clients at higher retainers.
- hire a person (as an outsourcer, or paid by the hour, or similar) to help you service the existing client first and the new clients later. Keep training that person so that there's less and less manual work for you, and keep overseeing everything she/he does (be the quality control person). This brings you closer to a real business (other people do the work, you sell and negotiate deals)
- let's say that you now have one colleague that does the work, and a couple of clients who pay just enough consulting to make you feel relatively safe. Let's say that you've organized this in a way that leaves you 50-75% free time to work on The Thing You Really Want To Do. Now, do you want to build a professional service business, or a product business?
In any case, if your work is the only source of income and you can't raise any capital (not even from friends, fools, and family), building recurring revenue streams first is how you get out of any rut, because getting out of the rut requires freeing up your valuable time first.
Does this make sense? Would any of this work in your case?
I'm definitely more interested in a product business, rather than a professional services business.
I've been trying to find small contracts and other sub contractors to push stuff on to. It's difficult to find clients who only need a small amount of work for the sort of things I do. Most of the potential clients I end up talking to either need something simple-yet-out-of-my-competence like throwing together a Wordpress theme, or they want to hire me full time for their early stage startup (at a huge pay cut) on a project that doesn't interest me. I might be bored, but I don't hate it so much that I'm willing to make life measurably worse in just about all aspects over it.
The few (two) potential clients who sounded ready to work evaporated after I mentioned putting together a contract. That's skeevy. Probably a good thing they disappeared.
And it's been extremely difficult finding developers who A) know what they are doing, and B) don't want full time work. I'm either finding great guys who want me to hire them full time (which I would love to do, if I had the money) or I'm finding chuckleheads who will take any scraps I can give them, but I have to redo all of their work in 2 to 3 months. My current client says I shouldn't hold everyone to my own standards. I don't know how to do that and my knee jerk reaction is that it's a terrible idea. I don't know how to make software with bad developers. I can't just sit by and watch Tweedle Dum and Tweedle Dee fail to complete tasks week after week that I know I could have gotten done myself if I wasn't spending so much time managing them.
I've been running side projects, but I'm blocked on turning "thing I'm excited about" into "thing that makes money". My most complete project so far is an ePub-generating, browser-based text editor. My friends use it regularly and love it, but it's been very difficult getting other users to it, so after about 5 months I think it's time to move on. I'm torn between just hunkering down and refining the tool even further + adding a pay wall, or leaving it the way it is and pursuing a new project with more potential.
I tried finding a PR/advertising partner. Everyone has their own projects and nobody wants to sign up for someone else's. I only looked for about two or three months, though. Perhaps I gave up early.
I think the big problem is that I am very isolated from people. I live in a suburb of DC. I moved here two years ago to be with my wife, a move made extremely easy by the fact that I was already freelancing and working from home. The only people I know down here are her friends. So I'm trying to get out to tech meetups, do some networking.
There is one task coming down the pipe for my current client that I'm somewhat interested in doing. I'm not sure I'm going to get to do it, though. He's been trying to hire someone with more experience in that particular area, plus there is a long list of other things that have to be done first.
In the past, I've been apt to just quit at this point and then figure out where things would go from there. I've usually been able to find a new job right away, as my skills are usually in demand, but it's always the same sort of consultoware that burns me deep. I'm trying not to follow the same pattern again. I wouldn't call it a mistake the times I've done it before, as it afforded me chances to see things from different perspectives and even got me into freelancing. When it was just me, not having any money to go on exotic vacations was fine. My wife has spoiled me in the last couple of years.
So I don't know. I feel like I've been spinning my wheels. I think I'll wait out my client to see if I get the chance to work the project I want, sock the cash away, and if I don&#...
And yes, networking helps. I have my own small "mastermind" group and the best thing I get out of it is inspiration, bursting my own bubble I have locked myself into, and decreasing fear of the future. I just wrote about it yesterday on my personal blog, so that part of your comment rang very near and very true to me.
In the end, you already know what to do :) I see from your comment that your direction is clear. Good luck with everything!
Along the way I was able to run a website that delivered the retail customer website of a billion dollar company using my code.
How something as silly as hosting helped make it happen..
I have hosted customer apps and sites in a datacenter since about 98. Networking, security was something that there was little choice to avoid picking up in addition to software development.
Forget about today, even 15 years ago (man it's weird typing that), hosting was quickly becoming outdated. Yet, there was still an earnest need that was going unfulfilled.
The need I see repeatedly is for complex/custom hosting of Web apps and websites instead of the basic ones.
Example today? Even something as simple as Wordpress is a pain to reliably host when there is traffic for the average person. Someone deciding to master WP has lead to a fantastic startup with WPEngine which sits on the premium end compared to it's peers.
This isn't for everyone: assuming you have the ability to develop your skills as needed, and with the right support, you can tackle your slice of the complex/custom/app hosting market.
Even small businesses with custom workflow or website apps often end up needing their own vps or dedicated server to maintain. If you're this passionate about hosting, I'm trusting that you have or are pursuing dedicated hosting skills.
Putting together a managed server hosting package that may or may not provide application level support can be quite stable income assuming the line is clearly drawn between code induced issues vs infrastructure induced issues.
How much is on the other side?
On the low end I have changed a few hundred a month, all the way up to a few thousand a month, so a customer can have a sys and app admin rolled into one.
The right kind of customers definitely have a peace of mind budget, where they want the discipline and consistent availability of someone who cares about them more than a contractor. The bottom rung of customers don't scale very easily, either.
The margins are decent IMO. You can get a decent sized VPS or dedicated server and easily have your costs under $2 per user per month. Then you charge the customer $9-29 per month.
They key will be automating the setup process. If you're doing traditional hosting, you may also need some sort of control panel (they all suck, btw).
We also sell other SaaS tools for photographers–allowing them to sell and share photographs. We upsell them to our website clients.
It's hard to define fairly easy to create recurring revenue. We were profitable from day 1 but it took more than a few years to clear $1M in annual revenue. And now there are a lot of well-funded competitors (wix, squarespace, etc.). So, my advice would be to find a niche, figure out what they need, and focus on them.
I have a few ideas (below). This is random, but I would advise you to avoid restaurants. I've tried it. Many others have too. They owners are too busy, have little money, and most just don't care that much.
Some other ideas I've had:
* A static website hosting service based on Jekyll. But a web-interface somewhere allows you to create new jekyll posts/pages.
* Wordpress hosting for landing pages. I like Unbounce but it's expensive. Create a WordPress theme with 12 different page styles and let me make an unlimited number of landing and lead-gen pages for it.
* Elementary school websites. As a parent of 4, I've yet to see a good one. I'm sure there are existing players, but if you can carve out a niche, there are COUNTLESS other things you could build for them. Start with some private catholic/christian schools near you. They have much less red tape in their buying process. If you have some sales chops, aim at the district level so you can bag a few schools at once.