Taking recurring screenshots of your product

Hi wonderful IndieHackers folks!

Over the years of being in the world of SaaS, I've always struggled to keep screenshots of our products up to date.

You spend ages putting together a beautiful marketing site, writing lovely support docs, and put screenshots of your product across all of them.

And then as soon as the product changes, your screenshots are out of date. Then you have to rinse and repeat the process.

Always takes a few hours of time, and is really dull and repetitive work – to a point where it's sometimes easier to just not bother.

Just wondering if others have found they struggle with this too?

Been toying with making a product from an internal tool we built to solve this – quick landing page I put together today to give more info: https://screenshow.io

Just really keen to see if others have had this issue or solved it in ways I am missing! Thanks all ✌🏾

  1. 5

    Back when I worked for a mobile app business, we used Fastlane to automate screenshots. Still, it was a pain to use.

    I haven't had to do this for web yet (still building my web app), but my initial reaction was that there should be a chrome extension to do this, but I couldn't find one. My second reaction would be to write a script in Selenium or Puppeteer, but this could be a time sink. So I think this makes sense as a product, but the hard part might be the infrequent usage. $0.50 / URL / month to me is weird pricing – I would want screenshots only when I update my app, not on some regular interval. I'm guessing that you're trying to create some saas recurring revenue, but I have a hard time paying a subscription for something that I use infrequently. Might make more sense as a one-time purchase of $50 or $100.

    1. 2

      I support this view, but I think if you could make the system charge only when it detects a change from the last set, you might be onto a winner. That way, high-frequency users will pay more to stay up-to-date, and more established products/sites will only pay when they make a change.

      Love the idea though. If only you could do it for video tutorials too! :oP

      1. 1

        Jaymie, this is REALLY handy to hear. And yes, triggering off, say, a build going out could be a really good approach, instead of a basic "ever month update" type flow.

        If we can ever get our heads around GPT-3 then maybe that's the day this can be applied to video tutorials?! I can dream!!

    2. 1

      Thanks so much for this valuable feedback, Steven!

      Yes on the pricing – ideally to make it financially viable I'll admit that recurring revenue would be nicer from a business point of view! But I totally understand the reluctance as a consumer of the product to commit to another recurring monthly expense unless it's a recurring monthly problem to solve ;-)

      It may be that $0.50 / URL / month is not the right price point / metric, so going to keep iterating here.

      1. 2

        Also, don't listen to me or anyone else on pricing – listen to the market. No one's going to say "I love paying subscriptions!" but the market may still support it. Maybe your target market is large businesses and they couldn't care less about paying a subscription... in general, I don't think indie hackers make the best customers.

        Also, another problem is that $0.50/url/mo is the worst of both worlds: subscription AND too low a price. Better off with something like $10 or $20 or $30/mo for a larger bundle.

        I used to work for a mobile app shop that made a mobile faxing app (send/receiving faxes from your iPhone). In faxing, there are many ways to skin the cat. Some competitors charged per page (e.g. 99 cents/page), some sold credits in bundles, some charged subscriptions. We sold credits in bundles to send faxes. 1 credit costs $3.99 and could send up to 5 pages. Then we bundled credits e.g. 10 credits cost $12.99, 20 credits cost $19.99. Many of our customers were one-time or infrequent users. But many were recurring users, buying credit bundles every single month. So even if the purchase was not a subscription, the revenue was recurring.

        Then we added the ability to receive faxes. We could have used credits or per page, but since we were renting out phone numbers, customers felt fine paying a subscription – since people already pay monthly for phone numbers, many customers felt fine with this idea. We had tiers like 14.99/mo for 100 pages, 29.99/mo for up to 500 pages. If you went over, you could upgrade to the next tier, or use credits only for the overage. Not surprisingly, we made much more money on subscriptions than credits. Anyway, my point is that there are lots of ways to experiment with pricing.

        Good luck!

        1. 1

          Thanks for this Steven – so much wisdom in this post.

          "No one's going to say "I love paying subscriptions!"" – never was a more truthful word spoken!

          Bundling up and making it a fixed monthly fee has come up consistently. I'll be honest the pricing is so far simply some words on a webpage and nothing has been fully defined yet, so I am listening to a lot of the feedback and using that to come to conclusions on what to ultimately define.

          Thanks again Steven – really appreciate the thoughts and you sharing your experience.

          1. 1

            Happy to help, and good luck!

  2. 4

    I'm not sure how comparable the features are, but there is already www.url2png.com www.stillio.com urlbox.io browshot.com screenshotapi.net and no doubt more.

    However all but Screenshot API have intro packages around the $30pm month mark, which is pretty steep for your use case of refreshing just a handful of URLs. I've not read to see who offers a login feature.

    Anyone with an automated testing setup leveraging headless browsers would easily be able to squeeze a few screenshots out of their existing setup (and without having to maintain a dummy account that your app stores credentials for)... however lets be honest, how many people would like to do more automated testing and how many are actually doing it :)

    The other players show there are customers. And since you've done a lot of the work already, I think it's worth a shot. A login feature, competitive pricing, slick presentation... why not?!

    1. 2

      Joejoejoe, this is sooo helpful, thank you! This is again why I LOVE the IndieHackers community. I wasn't actually aware of all of those other players in the space – and perhaps if I already was I would have been put off even suggesting the idea... But actually I do feel that the value proposition and ultimate execution here is different and CAN be different to those existing services to a point where they're not strictly viewed as competitors.

      And also, I suspect there's an opportunity to grow this market – it feels like, at least from conversations so far, that a lot of people are still struggling with this problem and there's still a lot of room to innovate and improve here.

      Thank you – hugely appreciate your support and positive energy!

      1. 2

        My pleasure. I'd actually thought about making a product like this before.

        It's a nice piece of tech to own, as it can be extended in various directions.

        Once you have a headless browser logging in, its not much more effort to fill some forms, press submit, toggle different UI elements open etc.

        Your error handling of this process would in effect be an uptime monitoring tool (better than ones get just GET the homepage which is usually a redirect to a cached login form anyway).

        You can get more advanced combinations of UI states in your screenshots, you can analyse differences in images (using something like https://github.com/rsmbl/Resemble.js) to notify of changes, spot errors etc.

        Lots of direction you could take your project instead or as well as the original one

        1. 1

          @joejoejoe LOVE these ideas – so much inspiration coming to mind here! Thank you so much – just loving the amount of support and energy in IndieHackers around this. Hugely appreciate it, thank you Joe!

  3. 4

    I think it's a really great and interesting idea.

    I wouldn't sell yourself short by putting this text on the very first page:

    "Why wouldn't I build this myself?
    Do you really want to spend your valuable time taking screenshots of your product every month? You have more valuable things to spend your time on. For just $2.50 / month you can have a screenshot of your homepage, pricing page, sign up page, features page, and about page – that's less than you'll spend on coffee this morning – even if you live outside of San Francisco."

    1. 1

      Thanks so much Kool – really appreciate you stopping by to give this feedback.

      Yes sometimes I feel when coming up with landing page copy, it's such a fine balance between pre-empting questions people have, and PUTTING questions / doubt in people's minds when they may not even have it!

      I'll take this on board – not sure it's a question too many people need answering on the homepage step, but maybe on a dedicated pricing page it'd be appropriate.

  4. 3

    Hi James! Nice to see you here (we've met long time ago during London Realtime Hackathon that you've hosted back in 2012).

    Following @joejoejoe reply, I think a better positioning for your service would be auto updating docs with up-to-date screenshots.

    When updating products and UIs, there are lots of follow ups to do like updating workflows, updating documentations, release notes ... etc.

    The workflow would be:

    • Each url will have a unique variable (could be the url itself).
    • Docs will render images from your service.


    # How to do X.

    And images will show the latest ones generated from your service. This will save companies more time than just having screenshots in a folder.

    1. 2

      @oras Hi Oras! Oh wow, thanks so much for this – great to be chatting again! Haha London Real-Time feels like a lifetime ago – great memory!

      I see what you are saying – so rather than storing images in a fixed location, when you are inserting images in a support doc, you simply call from a URL on our service that's always up to date.

      Thanks Oras – so good to be chatting again!

  5. 1

    Sounds interesting, but like @stevenkkim, I wouldn't want to pay a recurring fee for this.

    1. 1

      @dSebastien – thanks for the note, would you be happy paying anything for this? A fixed fee? If so, how much and what for?

  6. 1

    Lots of products are behind a login; have you thought of a solution for that?

    1. 1

      Thanks Stephan for the comment! Yes this is absolutely the number one question / request so far from everyone I’ve spoken to. We have solved this internally, so next need to figure out how to solve this on behalf of other customers. Really appreciate you raising this as a concern.

      1. 1

        Looks like a job for something like Cypress; authentication can be part of a scenario, and you can take screenshots/videos of each step. If you play with timing, and store the artifacts, then you have a good basis. With a cool UI on top to easily edit those scenarios and go through the collected screenshots/videos, it might be quite nice.

        1. 2

          @dSebastien Cheers for the thinking here – cool ideas, and I'll check out Cypress – hadn't come across it before.

Trending on Indie Hackers
Share your product or landing page, and I'll give you product design advice 95 comments What did you work on this weekend? (Jan 15-16) 23 comments How do I transition from a wantrepreneur to an entrepreneur? 22 comments Copywriting Examples — The world's best copy. In one place. 13 comments Building a microsaas in public 9 comments Working towards an MVP 9 comments